Independent software vendors and use cases
In this chapter, we describe various independent software vendors (ISVs) and use cases for the V9000 solution. IBM industry sales teams or IBM Business Partners and ISVs often assist customers in developing these solutions to find the best fit for their environments. FlashSystem V9000 offers significant capabilities to enhance any solution, as demonstrated in this chapter.
This chapter includes the following topics:
12.1 Use cases and ISV overview and considerations
The following topics describe real-world usage scenarios running with FlashSystem V9000:
•VMware and vCloud
•IBM Spectrum Scale
•Database Acceleration:
 – Oracle
 – MicroSoft SQL Server
 – IBM DB2
 – SAP
 – Epic
•Data deduplication:
 – Atlantis ILIO
 – Permabit
 – IBM ProtecTIER
•IBM Spectrum Control Storage Insights
•IBM Virtual Storage Center, which provides the functions of IBM Spectrum Control
 
Notes:
•For details about IBM Spectrum control, see the IBM Data Management and Storage Management web page:
•For details about the entire IBM Spectrum Storage family, see this web page:
12.2 VMware
VMware workloads including VDI can benefit from the extreme performance and macro efficiency attributes of IBM FlashSystem Family. Tight integration with VMware vSphere (including vCenter, VASA, and VAAI) helps drive up efficiencies and performance, and drive down complexity and cost. See the following YouTube video:
Many advantages exist to using IBM FlashSystem V9000 storage to run mission-critical workloads in VMware vSphere environments to help improver performance, reduce storage costs, and scale on-demand to support changing needs. For more information, see the following white paper, which describes the FlashSystem 840 environment:
This scenario is also described in the IBM Redpaper publication Deploying IBM FlashSystem V840 Storage in a VMware and Cloud Environment, REDP-5148.
FlashSystem V9000 extends this environment with advanced capabilities including VMware vCloud integration described in 12.7, “VMware vCloud integration” on page 482. As the evolution of these technologies continues, IBM and VMware continue to partner to extend these technologies.
 
Important: At the time of this writing, VMware vSphere 5.5 Update 2 and vSphere 6.0 introduced a new method for a datastore heartbeat leveraging Atomic Test and Set for heartbeat I/O.
Due to the low timeout value for heartbeat I/O using ATS, this can lead to host disconnects and application outages if delays of 8 seconds or longer are experienced in completing individual heartbeat I/Os on backend storage systems or the SAN infrastructure.
To roll back to the traditional method of heartbeat, issue the following command on each ESXi connected to IBM V9000 storage or the other Storwize-based arrays:
esxcli system settings advanced set -i 0 -o /VMFS3/useATSForHBOnVMFS5
This reversion of VMFS heartbeat activity is preferred rather than globally disabling VAAI or ATS when using applicable storage systems.
For more details and updates regarding this issue follow:
12.3 Database acceleration
FlashSystem V9000 provides enterprise customers an avenue to quickly and cost-effectively boost database performance for database infrastructures, for example Oracle, SQL Server, Systems, Applications & Products ( SAP), and IBM DB2. This section describes ISV scenarios that deployed FlashSystem storage with these database infrastructures that maximized performance with low latency, and elimination of performance bottlenecks, especially in OLTP performance, and implementation of FlashSystem V9000 advanced software functions. These advanced features include Real-time Compression, dynamic tiering, thin provisioning, snapshots, cloning, replication, data copy services and high-availability configurations.
For more details about IBM FlashSystem V9000 advanced software features, see Chapter 3, “Advanced software functions” on page 75.
12.3.1 Oracle
Oracle demonstrates an excellent example of how FlashSystem V9000 technology can be leveraged to implement solutions. By increasing efficiency in CPU utilization, the same workload can be performed on few licensed cores saving costs.
IBM POWER8 and FlashSystem accelerate Oracle
CTOs, CIOs, and system managers have a wide choice of server and storage technologies. This breadth of servers and types of storage can lead to decision-paralysis, where staying with existing technologies can seem to be the safest course of action.
Combining FlashSystem V9000 technologies with the new IBM Power Systems server with IBM POWER8® technology, outstanding performance can be demonstrated. In addition, the reliability and flexibility of the V9000 solution can provide significant implementation benefits.
IBM has documented performance characteristics of this solution using a brokerage workload to demonstrate the capabilities of the IBM POWER8 and FlashSystem combination in the following white paper:
The conclusion is that the Power System S824 server combined with IBM FlashSystem V9000 can offer a quantum leap forward in performance, scalability, and reliability. This combination of processing capability and reduced latency with increased storage capacity provides a new level of performance and capability unequaled by any other commercially available system.
 
Note: This paper was written about the FlashSystem 840, however the V9000 solution offers similar levels of performance, and includes advanced copy services technologies enabling the customer to implement robust solutions.
Oracle ASM and IBM FlashSystem leading practices
Oracle Automatic Storage Management (ASM) is an all-inclusive approach to storage management, performance, and availability. ASM is an excellent tool for managing mixed storage environments where both flash memory and hard disk drive (HDD) technologies are being used. It is an intrinsic block manager database that balances file access across disks and disk subsystems to eliminate hotspots and increase storage efficiency.
By striping extents across disk group members and mirroring across fail groups, ASM can give RAID performance and protection at the file-level of databases. ASM can also be used in conjunction with high-end disk arrays for mixed storage file management.
ASM overview
ASM operates as a lightweight Oracle database and offers the stability and reliability of a standard Oracle database. Therefore, multiple databases can be clients to a single ASM instance. This enables a single pool of disks to be efficiently used for various databases. In addition, ASM can be used in a RAC environment where ASM failover can occur nearly seamlessly between active nodes, permitting non-stop uptime for storage management.
ASM does not replace, but does complement disk subsystems and arrangements for efficient use in specific files. ASM will not automatically separate storage areas in disk groups for log files, archived backups, or database files. ASM templates assist in striping performance of each type of file, but sequential I/O is not differentiated from random I/O. It is still the duty of the architect to specify storage for specific roles. Sequential access log files should traditionally be held separate from the random I/O generated by database files.
This is not a flaw in ASM but a continuing practice of tiered storage that is driven by disk speeds. ASM is most effective when diskgroups are allocated with single homogeneous disks as their components.
Using high-speed IBM FlashSystem highlights the benefits of ASM. When multiple devices are available, the striping performance of ASM can linearly increase the throughput of the storage, reducing the response time to the client. In situations where extremely low-latency is needed, such as log writes, the effect of mirroring with ASM is negligible and the performance promises of a single IBM FlashSystem are met.
ASM features best used with IBM FlashSystem
ASM in Oracle version 11g includes advances that can take advantage of the performance of IBM FlashSystem. The features of preferred mirror read and fast mirror resync are the two most prominent features that fit in this category.
Preferred mirror read failure groups
The preferred mirror read is not a new idea, but is implemented in Oracle’s volume management in Oracle version 11g. The concept is to read from the storage that can present the needed data at a lower latency. Initially, this was designed for WAN or site-specific storage to avoid higher-latency site connections. By restricting data reads to the local storage, the application would be able to service requests at nominal read speeds, while writes were the only communication needed to traverse the long haul site link. This is a feature that is available to most SAN/disk managers and to Operating Systems with their included volume manager.
Fast mirror resync
Oracle version 11g includes the fast mirror resync feature, which tracks the changed extents for a given window of repair, a default value of 3.6 hours. If a disk goes offline, ASM will track the extents that are associated with the disk. When the disk is recovered and brought online (knowing that all previous data is still intact) the list of changed extents are applied to the recovered disk. This is extremely useful when working with very large volumes.
Conclusion
The high availability, performance gain and ease of management, lead ASM to a win for storage management. Preventing hot spots on disks intuitively leads to lower mechanical costs. Ease of management incurs lower operations costs. With higher performance, the fundamental cost per transaction is also reduced and by using an industry-leading database engine to serve storage, security, and availability are not compromised. FlashSystem V9000 enables Oracle administrators to use ASM to take advantage of extremely low latency when building solutions.
 
Note: See the white paper for Oracle, ASM, and FlashSystem:
Quick Oracle Statspack and AWR report analysis
Oracle Automatic Workload Repository (AWR) is used to collect, process, and maintain performance statistics for problem determination and self-tuning purposes. AWR is important to understanding FlashSystem V9000 benefits in the Oracle environment.
12.3.2 Microsoft SQL Server
Fast storage and low latency are vital to maximizing SQL Server performance. Adding servers or processing power often does not improve SQL performance because the system is limited by the speed of data access time. The result is a massive performance gap, often felt most by database servers, which typically process more I/O transactions than other systems. Super-fast processors and massive amounts of bandwidth are often wasted as storage devices take several milliseconds to access requested data.
IBM FlashSystem V9000 provides extremely fast storage performance with a small footprint, and also advanced software defined features in the base software including IBM Real-time Compression, thin provisioning, copy services, and a simple GUI that enables storage to be quickly deployed and easily managed.
IBM FlashSystem products offer extraordinarily high random I/O performance because of its low access times and high bandwidth. You can deliver more and have faster SQL transactions with FlashSystem storage.
Decreasing application performance under heavy user loads is not a new story for most enterprises. In the last three years, dramatic changes in demands have been placed on database servers. While the number of users of database systems has increased, so has the average amount of data stored in databases. Additionally, the demand for more complex business analysis has increased the complexity of the work done by database servers.
The combination of more users, greater volume of data, and more complex queries has frequently resulted in slower database response. Administrators often consider the following solutions for database performance problems:
•Server and processor performance: One of the first steps that most IT organizations do when performance decreases is to add processors and memory to servers or add servers to server farms. Adding servers and processors has a minimal effect on database performance that is storage bound, and compounds the resources that are wasted as even more processing power waits on the same slow storage.
•SQL statements: Enterprises invest millions of dollars squeezing every bit of efficiency out of their SQL statements. The software tools that assist programmers with the assessment of their SQL statements can cost tens of thousands of dollars. The personnel required for evaluating and iterating the code costs much more. In the last decade, many consulting firms have appeared that specialize in system tuning, and their number-one billable service is SQL tuning.
Tuning SQL can create performance improvements, but even the best SQL cannot make up for poor storage I/O. In many cases, features that rely heavily on disk I/O cannot be supported by applications. In particular, programs that result in large queries and those that return large data sets are often removed from applications to protect application performance.
In many cases, the money spent in these two pursuits can be significant, but the return is often disappointing. Server performance and SQL tuning alone do not often address the true cause of poor database performance; the gap between processor performance and storage performance. Adding servers and processors has minimal effect on database performance and instead compounds the resources wasted, as more processing power waits on the same slow storage.
Improving database performance
If you determine that your system has I/O subsystem problems, the next step is to identify which components of your SQL Server database are experiencing the highest I/O and in turn contributing the most to I/O wait time. Consider moving several components, such as transaction logs, temporary databases (tempdb), indexes, and frequently accessed tables to FlashSystem storage, or for simplicity and maximum benefit, you can even move the entire database to FlashSystem.
For more information, see the following resources:
•Faster Microsoft SQL Server Performance with IBM FlashSystem Storage, TIPS1038, an IBM Redbooks Solution Guide
•IBM FlashSystem Ecosystem solutions web page:
12.3.3 DB2
High performance, cost-effective, and easy-to-deploy solutions are available from IBM. IBM Power Systems provide the processing capabilities needed to scale out or scale up while powering the most complex analytics. IBM FlashSystem storage offers performance, efficiency and reliability, and also the virtualization tools that enable enterprises to turn the explosion of data into insight for business advantage. These IT infrastructure components provide the foundation upon which to build a next-generation data platform, using IBM DB2 that keeps transactional workflows and analytics operating at maximum efficiency.
Running with IBM Power Systems, which provide processing capabilities needed to scale out or up while powering the most complex analytics, IBM FlashSystem V9000 offers performance, efficiency, and reliability, and also virtualization and advanced software features that enable enterprises to turn the explosion of data into insight for business advantage.
Typical workloads
The following list describes some of the typical workload that clients run in DB2 environments:
•Brokerage workload: Examples might Include customer input such as trades, market analysis, and account inquiries.
•Transactional workload: Examples include transactions against a database, transactions that might include entering and fulfilling orders, accepting payments, inquiry of the orders status, and overseeing inventory levels.
•Big data insights analytics workload: Examples include online and catalog sales of merchandise.
Running applications on FlashSystem V9000 accelerates IBM DB2 workloads.
For more information, see the following resources:
•The IBM FlashSystem accelerates IBM DB2 workloads white paper at these addresses:
•Faster DB2 Performance with IBM FlashSystem, TIPS1041, an IBM Redbooks Solution Guide
12.3.4 Architecture for SAP landscapes featuring IBM FlashSystem
SAP provides a common centralized database for all the applications that are running in an organization. The database instance is a mandatory installation component for the installation of an SAP system.
SAP supports the following databases:
• Oracle
• MS SQL Server
• IBM DB2 Universal Database™ for UNIX and Windows
• SAP liveCache technology
• MaxDB
• IBM DB2 Universal Database for z/OS
• IBM DB2 Universal Database for System i®
• IBM Informix®
Although almost all storage systems can achieve high performance and high throughput with HDD, they need special technology to achieve low latency, especially for random I/O workloads with a small data block size. This is the typical SAP Business Suite workload.
Typically Flash technology has 10 - 50 times less latency than hard disk drives (HDD). By comparing one solid-state drive (SSD) with one HDD, or one SSD RAID with one HDD RAID, an SDD device performs faster than an HDD device, and Flash typically performs faster than an SSD. This is especially true for online transaction processing (OLTP), which is critical for SAP applications.
Integrating FlashSystem V9000 with its IBM FlashCore technology and advanced software technology with SAP and IT infrastructure helps address performance issues with business critical applications, and can either complement or replace traditional HDD and SSD arrays in OLTP or online analytical processing (OLAP).
IBM FlashSystem V9000 is a comprehensive, all-flash enterprise storage solution that delivers the full capabilities of IBM FlashCore technology plus a rich set of software-defined storage features, including the following features:
•IBM Real-time Compression
•Dynamic tiering
•Thin provisioning
•Snapshots
•Cloning
•Replication
•Data copy services
•High-availability configurations
FlashSystem V9000 advanced software functions benefit all virtualized storage. For example, Easy Tier optimizes use of flash memory. Real-time Compression enhances efficiency even further by enabling the storage of up to five times as much active primary data in the same physical disk space. Finally, high-performance thin provisioning helps automate provisioning. These benefits can help extend the useful life of existing storage assets, reducing costs. Thin provisioning helps improve efficiency by allocating disk storage space in a flexible manner among multiple users, based on the minimum space that is required by each user at any particular time.
The following list summarizes the advanced software functions, provided by FlashSystem V9000, that enhance an SAP environment:
•Improved utilization by pooling all physical storage capacity
•IBM Easy Tier for storage efficiency
•IBM FlashCopy for point-in-time copies in combination with IBM Tivoli Storage FlashCopy Manager
•High availability features through mirroring and copy services for data replication and protection
•Real-time Compression
IBM Easy Tier hot and cold data fully automated
Hot data is frequently accessed data, and the amount of data is too large to fit in any cache. Migrating it from HDD to SDD or to FlashSystem storage can provide significant performance improvement.
Cold data is less frequently accessed data or data that is not accessed at all. Migrating it from HDD to SSD might provide little or no benefit.
IBM Easy Tier enables more effective storage consolidation by eliminating the manual placement of data (files or file system) on flash and HDD. Easy Tier automatically and dynamically moves the appropriate data to the appropriate storage tier, based on ongoing performance monitoring. Easy Tier performs data relocation within SSD and flash memory and HDD storage pools (also referred to as a hybrid solution) to achieve the optimum performance, by moving the hottest extents to flash, then the cooler extents to SSD, and the coolest extents to HDD. It is done automatically, dynamically, and transparently to hosts, that is, without disruption to applications. This feature significantly improves the overall storage cost performance and simplifies the performance tuning and management.
Standard SAP operations eased with FlashSystem V9000
IBM FlashSystem V9000 virtualization and advanced software functions ease SAP operations and infrastructure administration.
From an operating system (OS) or server perspective only one type of storage device driver is required, delivered by any UNIX, Linux, or Microsoft based operating system.
Adding new or replacing existing storage is transparent to the OS and application, and can be done without any downtime. This is true for storage maintenance also.
Adding IBM FlashSystem or other types of storage to FlashSystem V9000 is transparent to the application, except that the application experiences a lower IO response time. No manual data, storage, or LUN configuration is necessary. This configuration is fully automated through IBM Easy Tier.
Database backup remains unchanged if database online backup is used, for example with IBM Tivoli Storage Manager for Enterprise Resource Planning (ERP), or can be enhanced and accelerated through FlashSystem V9000 FlashCopy in combination with IBM Tivoli Storage FlashCopy Manager supporting IBM DB2, Oracle, and MS SQL databases.
IBM Logical Volume Manager (LVM) is fully supported by SAP NetWeaver Landscape Virtualization Manager (SAP LVM), and simplifies operations, such as SAP homogeneous system copy and SAP copy or clone process. Even without SAP LVM, the IBM FlashSystem V9000 virtualization solution eases the process of copying SAP databases through FlashCopy; thin or full provisioned.
For more FlashSystem V9000 product details, see the IBM FlashSystem V9000 Product Guide, TIPS1281.
For more details about FlashSystem and SAP solutions, see SAP documents at the IBM FlashSystem Ecosystem solutions web page:
SAP full in-memory technology compared to classical implementation
SAP HANA is an in-memory, column-oriented, relational database management system developed and marketed by SAP SE. Architecture can handle both high transaction rates and complex query processing on the same platform. This solution can also include SAP HANA Tailored Datacenter Integration (SAP HANA TDI, formerly SAP High-Performance Analytic Appliance) within a SAN environment.
 
Note: For more information about SAP HANA TDI, see IBM System Storage Architecture and Configuration Guide:
In-memory technology delivers fast query results for Business Analytic workloads, and SAP HANA and also IBM DB0 BLU address increased performance.
As with every major change within an SAP landscape, for example, replacing the database from vendor A with vendor B, changing the server architecture from type X to type Y, there is a risk associated with it, and a price tag.
The following alternatives, which are standard and can be easy to implement, can improve performance with fewer or no risks, not changing the standard, established, operation model, but delivering significant performance improvements:
•Flash technology delivers much faster IO response time, and very low latency, compared to HDD. The integration of FlashCore technology with IBM is not apparent to SAP, so no operational mode is affected. Customers have experienced an SAP batch runtime improvement of factor 10.
•Upgrading and right-sizing the existing server to the latest technology level. Customers have experienced an SAP batch runtime improvement of factor 10, for example upgrading from IBM Power 5 to IBM POWER7® technology.
•Upgrading the database level to benefit from IBM in-memory technology with IBM DB2 10.5 BLU. This is a standard database upgrade. Customers have experienced SAP query time improvements by of factor 4 - 40.
Consider these three alternatives because implementing them is only an upgrade, not a replacement, and can lead to significant performance improvements, while keeping the architecture of the SAP infrastructure unchanged.
For more details, see SAP on the IBM FlashSystem Ecosystem solutions web page:
12.3.5 Epic
Epic is a set of software components that change the way healthcare interacts with patients. Epic covers all aspects of the healthcare organization, which directly affect patients, doctors and operations. This presents challenges to the infrastructure.
The ability of Epic to do what it was designed to do is what directly affects doctors and patients. As a result, solutions must meet minimum requirements when deployed in Epic environments.
Storage ensures that Epic can function efficiently and effectively. The processing ability of IBM POWER8 coupled with the enterprise capabilities and performance of IBM FlashSystem products accelerate performance needs that EPIC requires.
Unique storage challenge
In most use cases involving flash memory deployments, the business environment changes, driving a need for higher-performance storage. However, the case of Epic Systems Corporation software is the opposite. The storage requirements have not changed recently, but the options for addressing them have.
Epic is a privately-held company founded in 1979 and based in Wisconsin, US. It makes applications for medical groups, hospitals, and other healthcare organizations. Epic software typically exhibits high frequency, random storage accesses with stringent latency requirements. IBM works with Epic to develop host-side and storage-side solutions to meet these requirements.
Extensive testing has demonstrated that the combination of IBM POWER8 servers and IBM FlashSystem storage more than meets the performance levels that Epic recommends for the back-end storage supporting its software implementations (at a cost-point that is multiple times lower than other storage alternatives).
Award-winning Epic software
Epic’s integrated electronic medical record (EMR) software covers all aspects of healthcare operations, including clinical, billing, and revenue, in addition to patient record access. Epic software implementations employ two main database technologies:
•The OLTP database runs cache from InterSystems Corporation as the main OLTP database engine.
•Epic applications use analytical databases that run on Microsoft SQL Server or Oracle database software.
The analytical database component has the highest bandwidth but the cache OLTP database performance is by far the most critical to user satisfaction, and consequently is where most of the attention on storage performance has been focused.
The Epic solution is described in the following white paper:
The Epic and IBM POWER8 solution is detailed in the solution brief on the following site:
FlashSystem V9000 and Epic are a good match. Supplementing exceptional high performance with a complete set of enterprise storage services that include FlashCopy, cloning, replication, and compression, built on proven Spectrum Virtualize technology, makes the V9000 an excellent choice for Epic deployments.
12.4 IBM Spectrum Scale
IBM Spectrum Scale is a flash accelerated, industrial strength, highly scalable software-defined storage that enables global shared access to data with extreme scalability and agility for cloud and analytics.
Together, Spectrum Scale and FlashSystem enable enterprises to derive greater value from high-volume and high-velocity big data streams. The IBM FlashSystem V9000 flash memory arrays can help organizations harness the power of big data.
This combination helps ensure that critical applications operate with high availability and peak performance. Spectrum Scale is a proven, scalable, high-performance data and file management solution used extensively across multiple industries worldwide. FlashSystem storage arrays are data center-optimized to deliver extreme performance, reliability, Integrated deployments of FlashSystem and Spectrum Scale are enabling organizations to move away from traditional disk storage solutions cost-effectively. Three key use cases have emerged:
•Using FlashSystem as a cache device with Spectrum Scale
•Using FlashSystem to accelerate Spectrum Scale metadata
•Using FlashSystem as a storage tier dynamically managed by Spectrum Scale
Figure 12-1 shows an example of the IBM Spectrum Scale architecture and storage environment, including FlashSystem products.
Figure 12-1 IBM Spectrum Scale architecture
12.4.1 IBM Spectrum Scale benefits
Elastic Storage provides simplified data management and integrated information lifecycle tools capable of managing petabytes of data and billions of files.
As a type of clustered file system, Elastic Storage removes data-related bottlenecks by providing parallel access to data, eliminating single-filer choke points or hot spots. Elastic Storage is portable operating system interface (POSIX)-compliant, so organizations can implement enterprise-class, file based storage solutions without modifying their applications. Plus, Elastic Storage supports a wide range of file system block sizes and types to match I/O requirements.
IBM Spectrum Scale also performs important data storage and management functions, including information lifecycle management (ILM), disk caching, snapshots, replication, striping, and clustering. Therefore, organizations can store the correct data in the correct place at the correct time, with the correct performance and at the correct cost; automatically.
For more details about IBM Spectrum Scale see IBM Spectrum Scale (formerly GPFS), SG24-8254.
12.4.2 IBM FlashSystem benefits
IBM FlashSystem V9000 is a comprehensive all-flash enterprise storage solution that delivers the full capabilities of IBM FlashCore technology plus a rich set of software-defined storage features, including IBM Real-time Compression, dynamic tiering, thin provisioning, snapshots, cloning, replication, data copy services, and high-availability configurations.
As of FlashSystem V9000 software V7.5 support, extra functions are included, such as IBM HyperSwap capability, increased IBM FlashCopy bitmap space, Microsoft Windows Offloaded Data Transfer (ODX), and direct 16 Gbps Fibre Channel host-attach support (except for IBM AIX). Support also exists for VMware 6 that enhances and improves scalability in VMware environment.lashSystem storage arrays designed to deliver extreme performance, IBM MicroLatency, macro efficiency, and enterprise-grade reliability and serviceability.
FlashCore Technology
At the heart of FlashSystem V9000 is IBM FlashCore Technology, which consists of these key elements:
•Hardware-accelerated architecture that is engineered for flash, with hardware-only data path
•IBM MicroLatency modules that are designed for low latency, density, and reliability
•IBM Advanced Flash Management that improves MLC flash endurance 9x over standard implementations without sacrificing latency
To learn more about IBM FlashCore technology, see the following website:
IBM FlashSystem V9000 and IBM Spectrum Scale running together are a powerful, cost-effective solution for accelerating leading edge scientific, academic, governmental, and commercial computing environments where elastic storage is often deployed.
12.4.3 FlashSystem as a cache device with Spectrum Scale
Extremely fast, low-latency, high-bandwidth FlashSystem storage arrays are now certified for Spectrum Scale. FlashSystem provides the speed of internal flash, but with the flexibility and infrastructure of traditional shared storage solutions. Together, FlashSystem and Spectrum Scale can support millions of transactions without bottlenecks.
Many workload dynamics make it difficult to isolate which data should be on the highest performance storage tier. FlashSystem solves this challenge when deployed as a cache device used to accelerate the most active data. The hottest files are transparently copied to the FlashSystem cache and access to those files served at the speed of flash, rather than waiting on slower disk storage.
Because workloads shift, only the hottest data is kept as a fail-safe copy in the read cache of Spectrum Scale. This enables multiple workloads to use FlashSystem more efficiently and cost effectively.
12.4.4 FlashSystem for metadata acceleration for Spectrum Scale
Using FlashSystem to accelerate metadata (as a separate high-speed metadata broker), organizations can dramatically improve the performance of primary data workloads across Spectrum Scale environments. With the addition of FlashSystem all-flash storage media, metadata small-block I/O operations no longer interfere with the large streaming access operations of the primary workloads. Compute tasks such as batch processing and nightly backups have been significantly shortened by moving metadata stores to FlashSystem storage.
Spectrum Scale employs a distributed metadata architecture that enables exceptional scalability. Optionally, Spectrum Scale can be configured with dedicated metadata servers. FlashSystem improves the performance of this design. Especially for metadata changes that require synchronous updates, FlashSystem ultra-low latency for all I/O enables Spectrum Scale metadata processes to operate with minimal waits, even though metadata uses only a small percentage.
12.4.5 FlashSystem as a storage tier dynamically managed by Spectrum Scale
When using Spectrum Scale to create tiers of storage based on performance or other characteristics, FlashSystem can be deployed as a high-performance, flash-accelerated tier that helps improve the performance of the overall infrastructure. This increases efficiency and eliminates wasted capacity previously needed for performance. When data is moved between storage pools, the management is done automatically by Spectrum Scale.
FlashSystem enables effective storage tiering by offering the most cost-effective high-performance storage solution. To achieve the lowest storage costs, active data can be moved to FlashSystem, while inactive data can stay on the least expensive storage for raw capacity. This approach enables organizations to save money by not overspending on any storage tier for their application needs.
When it comes to performance, parallel file systems such as Elastic Storage are mainly limited by the metadata store, and this is exactly where FlashSystem offers the most value. In Elastic Storage environments, metadata usually comprises less than two percent of the capacity of the file system, but it is, of course, heavily involved in file system operations.
12.4.6 Use cases: Scientific and medical high-performance computing
This section describes a use case in a scientific environment.
Metadata accelerator
The Research Computing and Cyber infrastructure (RCC) group at Penn State University provides systems and services that are used extensively in computational and data-driven research and teaching by more than 5,000 faculty members and students. By deploying FlashSystem to accelerate Elastic Storage metadata stores, RCC administrators were able to increase overall system performance by 600 percent.
Workload accelerator
A large research hospital based in New York City is no stranger to the rigors, challenges, and rewards of medical research, backed by high-performance computational and data services. For example, the hospital’s medical students are allowed to fully sequence their own genomes, which requires extreme computational resources.
At the hospital, each genome processing project requires its own directory, often with more than a million files and 10 thousand folders, typically comprising 200 GB in total stored data. The sequencing application processing profile involves thousands of processors randomly accessing many small files. In this type of processing environment, HDD performance maxed out when handling the smaller data processing operations, leaving little room for larger operations.
The hospital deployed Elastic Storage to create tiers of production data on FlashSystem storage, including the raw genome data and files, common related scientific data and other reference sources. All of this data was moved by Elastic Storage into storage pools powered by FlashSystem. System IOPS escalated dramatically. As the IT administrators tuned block sizes and processing threads, they saw a significant increase in IOPS performance.
For more details, see the IBM white paper Deploy smarter, faster infrastructure with IBM storage solutions at the following website:
12.5 IBM Spectrum Control Storage Insights
IBM Storage Insights is an off-premise, cloud-based software as a service (SaaS) offering from IBM Service Engage. The IBM Storage Insights product is part of the IBM Spectrum Family of products and, as a Service Engage offering, its core runs over IBM SoftLayer®.
It combines IBM analytics leadership and a rich history of storage management expertise with a cloud delivery model, enabling you to take control of your storage environment.
When you add IBM FlashSystem V9000 storage, in addition to other IBM supported storage systems, you can monitor the usage of storage space and the performance of the storage systems.
IBM Storage Insights is designed to be easy to configure and to use. For more information about a trial, see the IBM Service Engage website:
The Storage Insights dashboard shown in Figure 12-2 provides an at-a-glance view of the entire storage environment, with the ability to drill down into efficiency-improving suggestions.
Figure 12-2 Storage Insights Home dashboard
12.5.1 Deployment
Only a single software component, the data collector agent, must be deployed to begin using Storage Insights. The data collector agent acts as a proxy agent that communicates to the storage devices, gathers information that pertains to configuration and performance, and sends it to the Storage Insights instance on the cloud.
Spectrum Control Storage Insights Data Collection
IBM Storage Insights can be deployed, be set up, and start to collect data in less than 30 minutes.
Gaining insights
To gain insights into planning capacity, storage tiering, reclaiming storage, and performance, view the charts and information that are provided about the storage systems that are being monitored:
•Viewing insight charts
Go to the pages that show the charts that provide overall views of the capacity, space usage, and performance of the storage systems that you added for monitoring. You can also view charts that show how space can be redistributed across tiers to improve performance, and show how to reclaim space by removing volumes that are not being used for storage.
•Capacity views
Gain insights into the capacity and the usage of space in your data center. You can see the capacity and space usage by storage system, by block storage, by file storage, and by application and by department.
•Performance views
Gain insights into the performance of the resources that are monitored in your storage environment. You can use performance metrics for volumes, disks, or ports to help you measure, identify, and troubleshoot performance issues and bottlenecks in storage systems.
•Reclamation views
Gain insights into the storage that you can reclaim. You can see the amount of storage that can be reclaimed in your data center, and in each of the storage systems that are being monitored.
•Tier planning views
Gain insights into optimizing storage in your data center. You can optimize the placement of volumes by implementing the recommendations to re-tier storage.
Supported Storage Systems
At the time of writing, IBM FlashSystem 900 and V9000 are supported by IBM storage solutions.
Figure 12-3 shows the supported storage systems.
Figure 12-3 Supported storage systems
See the Adding storage systems topic in the IBM Knowledge Center for the most current list of supported storage systems:
See the IBM Spectrum Control Storage Insights Product Page for more information for this easy-to-use and cost-effective Storage as a Service (SaaS) solution:
12.6 Data deduplication
This section describes the benefits of FlashSystem V9000 running with various data deduplication software and ISVs. Data deduplication can be defined as a specialized data compression technique for eliminating duplicate copies of repeating data. Various applications provide data deduplication techniques to improve storage utilization and can also be applied to network data transfers to reduce the number of bytes that must be sent. In the data deduplication process, unique chunks of data, or byte patterns, are identified and stored during a process of analysis. The following data deduplication software and ISVs running with FlashSystem products are described:
12.6.1 Atlantis ILIO
Atlantis ILIO is a server-side optimization software that offers data deduplication for virtual desktop environments at the source. This lowers the cost per desktop with better performance, further enhancing the economics of IBM FlashSystem V9000. Atlantis is proven at scale in the largest desktop virtualization deployments in the world and provides enterprise-class reliability with automated deployment and HA/DR.
Atlantis ILIO is a software that is deployed per physical host and integrates seamlessly with existing hypervisors and brokers. As the data deduplication is performed on the physical server, it reduces storage and network traffic, while increasing density and performance with MicroLatency of the IBM FlashSystem V9000. This makes persistent virtual desktops cheaper and faster than a physical PC.
Together, IBM FlashSystem storage and Atlantis ILIO optimization software deliver a unique solution that addresses the challenges of virtual desktop environments. The integrated solution delivers consistently better performance than a PC as projects scale, while reducing storage infrastructure requirements by up to 95%. Built on fully validated reference architectures, the FlashSystem-ILIO virtual desktop infrastructure (VDI) solution supports centralized management with flexible deployment options that can scale to tens of thousands of users.
For details of the cost-effective, high performance, and scalable VDI solution of running IBM FlashSystem products and Atlantis ILIO, see the IBM Systems Technology white paper Executive Summary, IBM FlashSystem and Atlantis ILIO:
Also see the IBM FlashSystem Ecosystem solutions web page:
12.6.2 Permabit
Permabit provides a high performance inline data deduplication and compression technologies for primary flash, hybrid, and HDD storage.
The Albireo SANblox is a 1U form factor system, which is paired for high availability (HA) and built on proven technology from Permabit, Emulex, and Red Hat Enterprise Linux. It features a simple web-based CLI and GUI for management and offers high performance inline data deduplication and compression. After the inline (4 KB) data deduplication and compression, the data is immediately written to block storage because the SANblox performs no write caching.
SANblox nodes are always deployed in pairs for seamless HA with transparent failover. For data deduplication SANblox analyzes each block to determine if it has been seen and redundant blocks are deduplicated. The remaining unique blocks are compressed, packed together, and stored on the FlashSystem V9000. This typically provides a 6:1 data reduction for most data types.
SANblox can be deployed with FlashSystem V9000 for a lower effective cost of flash memory.
Typical use cases are as follows:
•General-purpose storage
•Test and development (DevOps) environments
•databases (Oracle, SAP, SQL, MongoDB)
For more details about the benefits of running Permabit SANBlox appliances and IBM FlashSystem products, download the following IBM FlashSystem and Permabit FAQ:
12.6.3 IBM ProtecTIER
Data protection, in the form of backup, restore, and replication plays a key role in ensuring applications are always available. Products like IBM ProtecTIER have changed the landscape for data protection by implementing Enterprise class, scalable, and high-performance data deduplication backup solutions. ProtecTIER uses the patented IBM HyperFactor® data deduplication technology to deliver storage savings by a factor of 25 or more.
Backup, restore, and replication using ProtecTIER and FlashSystem technology delivers an optimized solution that offers the performance and throughput of flash, and advanced software features such as advanced software features. For a price that is half that of traditional hard disk solutions. Customers can also reduce their storage footprint by racks full of disk enclosures to a compact FlashSystem enclosure.
Making the solution even more cost-effective, FlashSystem can be implemented with ProtecTIER in several specific configurations, which enables flash performance to be targeted at the exact parts of the infrastructure requiring it.
Delivering high performance backup is all about achieving high throughput. In a ProtecTIER system, the data deduplication process operates inline (or in real time) to achieve best performance, generating both read and write activity. When data is restored from ProtecTIER to the primary system, the contents of the backup are read from storage and, because of the nature of data deduplication, results in highly random I/O.
FlashSystem arrays, which include MicroLatency, provide extreme performance with extremely low latency, making them particularly suitable for economical deployment with ProtecTIER.
FlashSystem storage can be used with ProtecTIER in several configurations to suit the requirements of the customer. These provide a balance between price and performance, depending on the capability required:
•Flash for metadata
FlashSystem can be deployed to host the metadata for one or more ProtecTIER systems. Metadata information is constantly accessed to store new content and retrieve backup data for delivery to the host. The user data can remain on spinning disk if that provides sufficient throughput. In some scenarios, it might be more cost-effective to share FlashSystem between two or more ProtecTIER clusters, enabling each to take advantage of metadata performance improvements.
•Flash for metadata and user data
FlashSystem can be used to host the entire ProtecTIER environment, covering both system metadata and user data. This scenario would be beneficial where a high volume of single stream backup and restore is required.
ProtecTIER with FlashSystem reduces the wasted space and expense seen with deploying large numbers of hard disks just to meet the throughput requirements of high performance ProtecTIER environments.
One other benefit of using FlashSystem for the data repository is the ability to encrypt data-at-rest on the array. If data is encrypted before being sent to ProtecTIER, any data deduplication benefits might not be realized. Encryption after the data is deduplicated is therefore much more desirable.
 
Notes: Consult with your IBM sales representative or IBM Business Partner for more information about deploying the IBM FlashSystem V9000 with IBM TS7650G ProtecTIER.
If the configuration that you want is not available for selection on the IBM System Storage Interoperation Center (SSIC) web page, submit a Solution for Compliance in a Regulated Environment (SCORE) to IBM, requesting approval. Another name for this procedure is request for price quotation (RPQ).
The IBM System Storage Interoperation Center (SSIC) is on the following website:
For more details about the benefits of IBM FlashSystem and IBM ProtecTIER, see the following PDF file:
12.7 VMware vCloud integration
VMware vCloud Suite is an integrated offering for building and managing a VMware vSphere-based private cloud. FlashSystem V9000 supports the vCloud environment with automation in several key areas.
IBM FlashSystem V9000 storage addresses business performance and capacity requirements, but a viable ISV integration that is needed by a software-defined environment (SDE) remains a challenge. The IBM Storage Integration Server addresses this requirement by bringing automation, elasticity, capabilities of storage as a service, and operations management for storage management.
 
Note: The FlashSystem product vCloud integrated solution in this topic describes IBM Spectrum Control Base Edition 2.1.1 and FlashSystem V9000.
The VMware vCloud Suite delivers virtualization across the data center enabling rapid provisioning applications with the correct levels of compliance and availability for improved efficiency, agility, and control. For more information, see the following web page:
12.7.1 FlashSystem V9000 in a VMware vCloud environment
This section provides an introduction for IBM FlashSystem V9000 storage with VMware and the IBM Storage Integration Server software, which delivers functions of IBM Spectrum Control. Figure 12-4 shows the current solution.
Figure 12-4 vCloud with FlashSystem V9000 integrated with IBM Spectrum Control Base Edition 2.1.1
Many integration points exist in VMware, including vCenter, vRealize Orchestrator, management capabilities in both traditional vSphere client and the vSphere Web Client or VASA. These VMware entities generate thousands of requests against the storage system to gather the information that they need.
Moreover, these entities are requesting the same information from the storage system. This approach, with many redundant requests against the storage system, does not scale well in very large environments that have multiple virtual centers.
In large environments, those numerous requests cannot be handled regularly in a timely manner. For example, when using the IBM System Storage plug-in for vCenter in such large environments, the time that is spent to refresh that agent (plug-in) can easily be 30 minutes or as much as one hour for very large sites. Using the traditional approach of having an agent (in this case, the vCenter plug-in) polling the storage system for each vCenter or ESX host is the cause of the long refresh time.
 
Note: At the time of this writing, VMware vCenter Operations Manager and VMware vRealize Orchestrator are not supported with version 2.1.1 of Spectrum Control. They are currently supported only by the IBM XIV storage array.
VVOL for FlashSystem V9000 is supported with VASA 2.
12.7.2 IBM Spectrum Control
The IBM Spectrum Control Base Edition 2.1.1 is a centralized server system that consolidates a range of IBM storage provisioning, automation, and monitoring solutions through a unified server platform. Version 2.1.1 replaces version 2.1.0 and fixes an issue with IBM Storage Provider for VMware VASA. The following solution components are included in the IBM Spectrum Control Base package:
•IBM Storage Enhancements for VMware vSphere Web Client (vWC) V2.1.1
•IBM Storage Provider for VMware VASA, V1.0 (rev. 2.1.0) and V2.0 (rev. 2.1.1)
•IBM Storage plug-in for VMware vCenter Orchestrator (vCO) V2.1.0
•IBM Storage Management Pack for VMware vCenter Operations Manager (vCOps) V2.1.0
For more information, see the IBM Spectrum Control Base Edition release notes and
user guide.
For more details about IBM Spectrum Control Base Edition 2.1.1, see the following website:
 
Note: IBM Spectrum Control provides efficient infrastructure management for virtualized, cloud, and software-defined storage to simplify and automate storage provisioning, capacity management, availability monitoring, and reporting.
The functionality of IBM Spectrum Control is provided by IBM Data and Storage Management Solutions, and includes functionality that is delivered by IBM Virtual Storage Center, IBM Tivoli Storage Productivity Center, and others.
This application facilitates the integration of Storage System resources by using options (user credential, storage system, and storage pool management) with supported independent software vendor (ISV) platforms and frameworks. It provides a foundation for integration with IBM systems and ISV solutions.
IBM Spectrum Control version 2.1 provides a single-server backend location, and enables centralized management of IBM storage resources for different virtualization and cloud platforms, including the following platforms:
•VMware vCenter Server
•VMware vSphere Web Client
•VMware vSphere APIs for Storage Awareness (VASA)
•VMware vCenter Operations Manager (vCops)
•VMware vRealize Orchestrator
•VMware virtual volumes (VVOLs)
 
Through unique credentials, storage array, and storage pool management options, the Spectrum Control server facilitates the integration of IBM storage system resources with the supported virtualization and cloud platforms, such as VMWare. The Spectrum Control server also provides a foundation for integration with future IBM systems and ISV cloud solutions.
It can be managed through a standard web browser and a graphical user interface (GUI), or through a terminal and a command-line interface (CLI).
Value to customers
Use of the IBM Spectrum Control creates an abstraction layer between the storage system and the operating systems. This abstraction reduces the complexity of storage services in a cloud environment by focusing on the business and workload goals of IT storage infrastructure and delivering the following customer values:
•Discover data store relationships down to IBM storage volumes and file shares
•View native IBM storage array, pool, and volumes and file share properties
•Self-provision volumes and file shares from delegated pools
•View IBM storage capabilities
•Optimize VM storage placement automatically (VMware Storage DRS)
Features
The following solution components, referred to as cloud interfaces, are included and supported by V9000 in the IBM Spectrum Control Base Edition V2.1 software package:
•IBM Storage Provider for VMware VASA
The IBM Storage Provider for VMware VASA improves the ability to monitor and automate storage-related operations on VMware platforms.
•IBM Storage Enhancements for VMware vSphere Web Client
The IBM Storage Enhancements for VMware vSphere Web Client integrate into the VMware vSphere Web Client platform and enable VMware administrators to independently and centrally manage their storage resources on IBM storage systems.
Where to download IBM Spectrum Control Base Edition V2.1
You can download the current version at no initial charge from the IBM Fix Central web page:
12.7.3 vCloud Suite
VMware vCloud Suite is a vSphere private cloud. VMware vCloud Suite enables IT to build and manage a vSphere–based private cloud, resulting in strategic IT outcomes. It does this by assembling an integrated set of products, engineered to work better together, which provide these benefits:
•Infrastructure virtualization
•Disaster recovery and testing automation
•Cloud management for on-premises vSphere environments
VMware vCloud Suite contains these integrated products:
•Infrastructure Platform
 – VMware vSphere: Industry-leading server virtualization platform
 – Disaster Recovery Automation with VMware vCenter Site Recovery Manager: Policy-based disaster recovery and testing for all virtualized applications
•Cloud Management Platform
 – VMware vRealize Operations: Intelligent performance, capacity, and configuration management for vSphere environments
 – VMware vRealize Automation: Self-service and policy-based infrastructure and application provisioning for vSphere environments
 – VMware vRealize Business: Automated costing, usage metering, and service pricing of virtualized infrastructure for vSphere environments
12.7.4 FlashSystem V9000
FlashSystem V900 data services are provided through FlashSystem V9000 software, which offers a rich set of software-defined flash features that include IBM FlashCopy, thin provisioning, remote mirroring, external virtualization, IBM Easy Tier 3rd Generation, and IBM Real-time Compression.
FlashSystem V9000 merges IBM software-defined storage with the scalable performance of IBM FlashSystem technology to help you more easily manage your entire storage environment, while preserving your investments in storage.
With software-defined storage services, you can use the following features across all of
your storage:
Thin provisioning Enable dynamic growth. Purchase only the storage you need when you need it.
Easy Tier flash memory management Optimize performance at lower overall cost.
High availability configurations Enable near-continuous data availability.
Copy services Create space-efficient backups.
Disaster recovery techniques Validate business continuity plans.
Simple GUIs Quickly deploy and efficiently manage storage.
External virtualization Leverage SAN storage investments by using existing capacity, tiering to or from it, to maximize your existing storage investment.
HyperSwap Manage disaster recovery. HyperSwap provides highly available volumes accessible through two sites at up to 300 km apart, while maintaining a fully independent copy of the data at each site.
 
The FlashSystem V9000 software provides a single interface into the storage environment.
For more details about FlashSystem V9000 see Chapter 2, “FlashSystem V9000 architecture” on page 19.
12.7.5 Use case: Provisioning FlashSystem V9000 volumes using VMware
This section has an example of a vCloud solution that is focused on VMware Site Recovery manager.
The IBM Storage Enhancements for the VMware vSphere Web Client plug-in is used to create and manage FlashSystem V9000 volumes in storage pools that are attached to the IBM Spectrum Control Base Edition server 2.1.
Figure 12-5 shows you how to create new FlashSystem V9000 volumes directly from vCenter.
Figure 12-5 Volume creation
Complete the following steps to use the plug-in to create new FlashSystem V9000 volumes:
1. Through the plug-in, select WIN-70507KRD4O (Figure 12-6). Then select Actions → All IBM Storage Enhancements for VMware vSphere Web Client Actions → Create New Volume.
Figure 12-6 The IBM Storage Enhancements for VMware vSphere Web Client plug-in
2. The Create New Volume window opens (Figure 12-7). The Host Mapping selection field shows that ITSO-Cluster is the selected host to be mapped to the newly created volumes. In this example, two 500 GB volumes are created and mapped to ITSO-Cluster.
Figure 12-7 Create mirrored volume
The initial volume is named ITSO-Cluster_Vol_1 and is created in the Pool1 on the perfV9000a storage array. The number in the brackets sequentially increases by one.
The second volume is called ITSO-CLuster_Vol_2. The Storage Array/Pool drop-down menu lists the storage pools that were configured on the IBM Spectrum Control Server.
Figure 12-8 on page 489 shows the additional volume properties that can be selected when you create a volume from the Storage Enhancements for VMware vSphere Web Client plug-in. The following volume properties are available to select:
 – Enable Thin Provisioning
 – Enable Data Compression
 – Enable Virtual Disk Mirroring
 
Note: Data compression requires the volume to be thinly provisioned. If you select data compression, thin provisioning is automatically selected for the volume. The thin provisioning selection is disabled, and a check mark is placed in the selection box.
Figure 12-8 Select additional volume properties
3. During volume creation, a mirrored copy of the volume can also be defined. When Enable Disk Mirroring is selected, a secondary pool and I/O Group must be provided. The Secondary Pool drop-down menu lists pools that are available on the storage array. We selected the secondary pool, Pool2.
4. The I/O Group drop-down menu lists the I/O groups that are available on the storage array. Select the io_grp0 I/O group.
Your selections here determine where the mirrored volume is placed on the storage array. Having a second pool on a second storage controller is a preferred practice for redundancy. These pools should have similar performance characteristics.
Figure 12-9 shows the new FlashSystem V9000 volumes, created and mapped directly from the vSphere Web Client, without the need for the VMware administrator to access the FlashSystem V9000 GUI or CLI directly. This view lists the storage array where the volumes are located. The names and size of the volumes can also be seen in this view.
Figure 12-9 FlashSystem V9000 volumes created and mapped from the vSphere Web Client
The results of the volume that was created by the VMware vSphere Web Client plug-in can also be seen in the FlashSystem V9000 GUI (Figure 12-10 on page 490). This particular view of the GUI lists the volumes that were defined on the FlashSystem V9000. The figure also shows the two volumes that were created from the plug-in. These volumes are named ITSO-Cluster_Vol_1 and ITSO-Cluster_Vol_Cluster_2.
Figure 12-10 FlashSystem V9000 GUI view, Volumes by Pool window Use cases: Availability solutions
In this section, we highlight two availability solutions of FlashSystem V9000:
•A single-site high availability (HA) solution using volume mirroring
•Cross-site (extended distance) solutions: metro and global mirroring, and global mirroring with changed volumes
12.7.6 Single-site HA: Volume mirroring to another storage system
Replication of data within a single site is typically used to protect against failure of a single subsystem by mirroring the data to another subsystem, usually within the same site. FlashSystem V9000 supports volume mirroring, which maintains a synchronous copy of the volume within the same storage pool or in different storage pools. Typically, the two copies are allocated in different storage pools. By using volume mirroring, having two copies of a volume provides a basic RAID-1 function at the storage level.
The first storage pool (in our example Pool1) contains the primary copy. The second pool, Pool2, contains the secondary copy. If one of the storage pools fails, a volume copy is not affected if it was placed in a different storage pool. Volume mirroring is apparent to the host. If one of the subsystems fails, the host continues without any effect. After the failed subsystem is repaired, the mirrors must be re synchronized.
Figure 12-11 shows volumes by pools that are defined on the FlashSystem V9000. The ITSO-Cluster_Vol_1 has two copies: Copy 0 and Copy 1. The asterisk (*) indicates that Copy 0 is the primary copy of the volume.
Figure 12-11 Mirrored volumes
When a server writes to a mirrored volume, FlashSystem V9000 completes the writes to both volumes. Similarly, when a server reads a mirrored volume, FlashSystem V9000 reads the copy from the primary volume. In the case where one of the copies is in flash memory and the other copy is on a spinning disk, a preferred read configuration can be implemented.
In a preferred read configuration, the copy in flash is typically defined as the copy used to satisfy all read requests from a host or application to the mirrored volume. The write performance in a preferred read is determined by the I/O performance of the slowest device. This provides the highest read performance back to the application. This configuration is typically used in database environments, but it can be implemented for any workload where you want to enhance the read performance.
12.7.7 Cross-site HA: Extended distance
If a greater level of protection is required, such as protection against fire or flooding at the primary site, moving the auxiliary storage to another location is necessary. FlashSystem V9000 provides additional copy services to protect data by creating a copy of the data at the secondary site. FlashSystem V9000 can replicate data to another FlashSystem V9000 at another location. It can also replicate to an IBM Storwize SAN Volume Controller or Storwize V7000 at a separate location.
FlashSystem V9000 offers the following copy services:
•FlashCopy creates a point-in-time copy and gives the application continuous operation during backups.
•Metro Mirror is a synchronous remote copy function. In a synchronous mirror relationship, updates are committed to both the primary and secondary copy before the host I/O completes. This slightly increases latency, depending on the distance and type of technology used as the link between the two sites. It is suitable for disaster recovery solutions for limited distances. Metro Mirror is supported up to 300 km. It keeps consistent and current images of primary volumes.
Figure 12-12 shows a simplified view of a Metro Mirror relationship.
Figure 12-12 Metro Mirror
•Global Mirror offers an asynchronous copy function for distances greater than 300 km. Global Mirror is an asynchronous process, so the host I/O completes after the primary copy is updated. The update is then queued to the remote subsystem to update the secondary copy. This prevents the host from seeing any additional latency from delays on the remote link, but it means that the remote copy is always slightly outdated.
Global Mirror keeps consistent images of data at all times, even when the sites are separated by long distances. To maintain integrity, FlashSystem V9000 ensures that all remote updates are sent in the order that they are received from the host. Global Mirror supports up to 80 ms of round-trip latency.
Figure 12-13 shows the relationships between FlashSystem V9000 and Global Mirror.
Figure 12-13 Global Mirror relationship
•Global Mirror with Change Volumes is an asynchronous function that is based on adjustable point-in-time copies of data. Global Mirror with Change Volumes was designed for replication over lower-bandwidth networks. Figure 12-14 shows a simplified Global Mirror with Change Volumes relationship.
Figure 12-14 Global Mirroring with Change Volumes relationship
VMware offers Site Recovery Manager (SRM) as an optional element of VMware Infrastructure to enable automated disaster recovery. A Storage Replication Adapter (SRA) for the SRM is the link between VMware SRM and FlashSystem V9000. It makes VMware software fully aware of FlashSystem V9000 features, such as FlashCopy and Remote Copy. It gives the SRM the possibility to automate a recovery process by using FlashSystem V9000 storage replication services.
The benefit of adding SRM is that it enables VMware to manage the remote copy function of the FlashSystem V9000 and also the VMware hosts and guest virtual machines. It aids in the following tasks:
•Setting up the recovery infrastructure
•Creation of recovery plans
•Testing recovery plans
Automating failover
The combination of VMware Site Recovery Manager, Storage Replication Adapter, and FlashSystem V9000 enables the automated failover of virtual machines from one location to another location that is connected by IBM Metro Mirroring or Global Mirroring technology. This ensures that, if there is a disaster at the primary location, VMware vCenter is able to restart all of the required virtual machines at the secondary location with minimal delay and manual intervention and the shortest recovery time. For more information about the SRM, see the “VMware vCenter Site Recovery Manager Documentation” page at the VMware website:
12.8 Running FlashSystem V9000 in an IBM Virtual Storage Center environment
IBM Virtual Storage Center (VSC), a component of IBM Spectrum Control, provides an end-to-end view of storage, which helps you manage your data with more confidence. VSC can automate a wide variety of storage management tasks (including virtualization and provisioning) that enable storage to easily be integrated into a cloud computing environment.
 
IBM Spectrum Control: Provides efficient infrastructure management of virtualized, cloud, and software-defined storage to simplify and automate storage provisioning, capacity management, availability monitoring, and reporting.
The functionality of IBM Spectrum Control is provided by IBM Data and Storage Management, solutions, and delivered by IBM Virtual Storage Center (VSC).
For more information, see the IBM Data Management and Storage Management website:
IBM FlashSystem V9000 and IBM VSC provide efficient virtualization, management, and data protection for heterogeneous storage environments. This solution helps IT storage managers migrate to an agile cloud-based storage environment and manage it effectively without having to replace existing storage systems.
This powerful offering removes the physicality of storage, and also the complexity that is associated with managing multivendor infrastructures. It delivers to customers, under one licensed software product, the complete set of functions that are available in the IBM Tivoli Storage Productivity Center, the functions, and capabilities that are associated with the FlashSystem V9000 (including copy services), and the capabilities of the IBM Tivoli Storage FlashCopy Manager. This powerful solution enables organizations to optimize provisioning, capacity, availability, reporting, and management for virtual storage.
IBM Virtual Storage Center, a component of IBM Spectrum Control, offers these features:
•A single console for managing all types of data on disk, flash, file, and object storage systems.
•Analytics-driven tiered storage optimization that automatically moves data to the most cost-effective tier.
•Software-defined storage that is dynamic, service-oriented, and cost-effective. In this demonstration, you provision, monitor storage activity, analyze and optimize, and transform your storage resources according to their utilization.
As shown in Figure 12-15, IBM Virtual Storage Center includes core functionality, from three IBM offerings:
•Storage management through IBM Tivoli Storage Productivity Center
•Storage virtualization with FlashSystem V9000
•Application-aware data protection with IBM Tivoli Storage FlashCopy Manager
Figure 12-15 Overview of IBM Virtual Storage Center
 
Note: As of version 5.2.6, IBM Tivoli Productivity Center, a component of IBM Spectrum Control, is fully supported with FlashSystem V9000. See the current support matrix for more details:
FlashSystem V9000 with IBM VSC offers the following features in forming a storage cloud solution:
•Virtualization. VSC virtualizes heterogeneous storage at a block I/O level regardless of the vendors.
•Optimization. VSC optimizes the use of the block storage, creating a unique physical pool from which the customers can create their own tiers of storage.
•Analytics. VSC provides advanced storage analytics capabilities to discover, monitor, and report about the disk capacity usage and performances, and implement the chargeback; it simplifies the root cause analysis of performance issues.
•Policy automation. With VSC, customers can implement standardized storage provisioning by automation, with policies.
•Copy Management. Clients can use VSC to produce and manage multiple point-in-time copies of the data for backup, which can be sent to Tivoli Storage Manger for backup to media.
•Improved storage utilization. Make better use of existing storage and controlling storage growth expenditures.
•Improved application availability and simplified data migrations. Clients can make changes to storage and move data without taking applications down.
•Simplified storage management. VSC offers greater efficiency and productivity for storage management staff.
•Network-based replication and software architectural advantages. Helps enable greater choice (lower cost) when buying storage and lowers software costs.
•Reduction in storage management and administration cost. A core group of administrators can control multiple assets across a distributed storage environment (efficiency is improved by 50%).
•Improved storage utilization. Improves capacity utilization of existing storage assets.
•Controls future spending. Controls the growth of future spending (improves use by 30%).
•Reduced cost of storage. Capitalize on being able to purchase the lowest cost storage resources (growth is controlled, on average, by 20%).
•Improved customer and user availability to data-driven applications. Minimizes downtime associated with migrating data between storage assets.
See the following resources for more information:
•For more information about the implementation of VSC with FlashSystem V9000, see IBM SmartCloud Virtual Storage Center, SG24-8239
•The Executive Brief: Cost/Benefit Analysis of IBM Virtual Storage Center Compared to EMC Storage Virtualization Solutions:
12.8.1 Licensing
VSC for the Storwize Family (5608-ACL) can also be used for the FlashSystem V9000 internal storage enclosures. You must use VSC Standard (5608-AE1) for V9000 externally virtualized storage. If VSC is not required, you can continue to license FlashSystem V9000 external virtualization (5639-FV7) per enclosure for externally virtualized storage.
For the storage that is internal to the FlashSystem V9000 (meaning not storage from another vendor, which the V9000 might be virtualizing), you can license with the VSC for Storwize Family license (5608-ACL), licensed on a per storage device (per enclosure) basis. You also have the option to license that storage internal to the FlashSystem V9000 by terabyte capacity using the standard VSC (per terabyte) license (5608-AE1).
 
Note: The VSC for Storwize offering can be used on only storage that is internal to the FlashSystem V9000. For external virtualization, you must use VSC (the standard, per terabyte offering, 5608-AE1). If VSC is not required, you can continue to license FlashSystem V9000 external virtualization (5639-FV7) per enclosure for externally virtualized storage.
The following criteria must be completed according to your business environment:
•IBM FlashSystem V9000 software (5639-FS7) for all of the FlashSystem V9000 storage enclosures that are internal to the system have been purchased
•IBM SmartCloud® Virtual Storage Center (5608-AE1) or IBM SmartCloud Virtual Storage Center for Storwize Family (5608-ACL) licenses that have been purchased for all internal (5608-AE1 and 5608-ACL) and external enclosures (5608-AE1 only), managed by FlashSystem V9000 system
Organizations receive the following entitlements:
•The complete set of advanced functions available in IBM Tivoli Storage Productivity Center Storage Analytics Engine
•All the capabilities of IBM Tivoli Storage Productivity Center
•All the capabilities of IBM Tivoli Storage FlashCopy Manager
•All of the functions available for virtualization and remote mirroring
 
Note: At the time of writing, IBM Tivoli Productivity Center, a component of IBM Spectrum Control, is not yet supported with FlashSystem V9000. For details about the current Tivoli Productivity Center and storage support, follow the current support matrix for details:
 
..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset
18.191.78.136