Home > Workload Solutions > High Performance Computing > White Papers > Dell Validated Design for Health Care and Life Sciences with Dell PowerEdge C6520 and XE8545 > Performance evaluation of NVIDIA Clara Parabricks
Clara Parabricks germline pipeline 4.0.0-1 is tested using PowerEdge XE8545 with four A100 GPUs (SMX4, NVLink version). The results with two A100s are compared from version 3.6.1-1 with PowerEdge R7525 for a fair comparison. The main difference between the two server configurations is the number of GPUs and NVLink as shown in Table 2. Figure 11 shows the runtime difference with two A100s between two different servers with and without NVLink support. The runtime reductions between two servers with the Clara Parabricks version 3.6.1-1 are 16%, 32%, and 13% for 10x, 30x, and 50x WGS data, respectively. There is no notable performance boost from version 3.6.1-1 to version 3.7.0-1 and from version 3.7.0-1 to version 4.0.0-1 on the identical system. These observations are confirmed with the Clara Parabricks team, that version 4.0.0-1 is focused on the inclusion of new functionality, and was not focused on additional accelerations.
NVIDIA continues to introduce software improvements to Clara Parabricks. The latest version, 4.0.0-1, focused on adding more tools than improving performance as shown in Figure 12. Dell Technologies has observed continuous performance improvement from an older version to a newer version.
The minimum number of GPUs required for Clara Parabricks is two. As shown in Figure 13, the runtimes scale well for 2x and 4x GPUs with various sizes of WGS data. The previous test results with NVIDIA T4 GPUs show linear scalability up to 12 GPUs with 50x WGS data.