A specification called BLAS is commonly used in order to perform linear algebra operations. This library was originally created as a FORTRAN library in 1979 and has been maintained since then. BLAS has many optimizations for performant manipulation of matrices. Because of the depth and breadth of this specification, many languages have chosen to use this specification as part of their linear algebra libraries within their domain. The Go Sparse library uses a BLAS implementation for its linear algebra manipulation. The BLAS specification is composed of three separate routines:

• Level 1: Vector operations
• Level 2: Matrix-vector operations
• Level 3: Matrix-matrix operations

Having these leveled routines helps with the implementation and testing of this specification. BLAS has been used in many implementations, from Accelerate (macOS and iOS framework) to the Intel Math Kernel Library (MKL), and has been an integral part of linear algebra in applied computer science.

Now, it's time to learn about vectors.