White Papers
17 Dell HPC NFS Storage Solution – High Availability (NSS7.0-HA) Configuration
5 NSS7.0-HA I/O Performance
This section presents the results of the I/O performance tests for the current NSS-HA solution. All
performance tests were conducted in a failure-free scenario to measure the performance of the solution.
The tests focused on two types of I/O patterns: large sequential read- and write operations, and small
random read- and write operations.
The 480 TB configuration was benchmarked with Intel OPA cluster network connectivity. The 32-node
compute cluster described in section 4.2 is used to generate workload for the benchmarking tests. Each
test was run over a range of clients to test the scalability of the solution.
The IOzone tool was used in this study, which was used for the sequential and random tests. For
sequential tests, a request size of 1024 KiB was used. The total size of data transferred was 512 GiB to
ensure that the NFS server cache was saturated. Random tests used a 4 KiB request size and each client
read and wrote a 4 GiB file. Metadata tests were performed using the mdtest benchmark and included file
create, stat, and remove operations. All of the test results were based on NFSv4. For a list of commands
used in the tests, see Appendix A later in this technical white paper.
5.1 OPA sequential writes and reads
In the sequential write and read tests, the I/O access patterns are N-to-N. That is, each client reads and
writes to its own file. Iozone was run in clustered mode and one thread was launched on each compute
node. As the total transferred data size was kept constant at 512 GiB, the file size per client varied
accordingly for each test case. For example, 512 GiB file was read or written in one-client test case, 256
GiB file was read or written per client node in two-client test case.
Figure 3 shows the sequential write and read performance. The figure shows the aggregate throughput
that can be achieved when a number of clients are simultaneously writing or reading from the storage
over the Intel OPA network fabric.
During the benchmark tests, we observed that
The max read performance was up to 6.57 GB/sec
The max write performance was up to 1.84 GB/sec.