Dell PowerEdge EL Manual

Also see for PowerEdge M420: Manual Command reference guide Owner's manual Administrator's guide Information update

Contents

Dell PowerEdge EL Manual Manual pdf 17 page image

Improving NFS Performance on HPC Clusters with Dell Fluid Cache for DAS17• DFC in Write-Back mode (DFC-WB) – This configuration builds on the baseline by adding DFC asdescribed in Section 2.2, and DFC is configured to operate in Write-Back (WB) mode. WB modeallows the caching of writes on the cache pool. WB mode requires the data to be written to aminimum of two PCIe SSDs. Both re-reads and writes are accelerated.• DFC in Write-Through mode (DFC-WT) – This configuration builds on the baseline, but here DFCis configured in Write-Through (WT) mode. WT mode forces writes to both the cache andvirtual back end disk simultaneously.The following sections present the results of the different I/O patterns.3.1. Sequential writes and readsThe results of the IPoIB sequential write tests are shown in Figure 5. The figure shows the aggregatethroughput that can be achieved when a number of clients are simultaneously writing to the storageover the InfiniBand fabric.The results show that baseline configuration can reach a peak write throughout of ~2,000 MiB/s. Recallthat this is with the “sync” NFS export option, and this peak throughput demonstrates how well theconfiguration is tuned 4. With DFC in WB mode, labeled DFC-WB in Figure 5, the throughput measuredwas ~400 MiB/s with a peak of ~600 MiB/s. The lower sequential write performance with DFC is due totwo factors: the pure sequential write performance of the SSDs is lower than the storage array and thewrite-back cache has to replicate dirty blocks which means every write to the cache has to be writtento two SSDs. Subsequent re-write operations were found to have ~25% higher throughput as the filesare already in the DFC cache.Switching to WT mode and eliminating the replica blocks doubles the sequential write performance.Peak performance approaches ~1,000 MiB/s as seen in Figure 5, labeled DFC-WT. Recall that all writesgo directly to the backend disk in addition to being cached on the SSDs in WT mode.If the I/O pattern is such that there is a large amount of sequential data written to the backendstorage initially followed by subsequent reads, re-reads, and small writes, one method to takeadvantage of DFC performance on random workload (Section 3.2) while minimizing the sequential writeperformance penalty is to disable caching on the backend disk during the write operation. Although thismight not be an option for production clusters, smaller single-user/single-application environmentsmight be able to adopt this approach. DFC provides very simple utilities to accomplish this. Once thedata is written to the backend store, for example by a gene sequencer, caching on the backend diskcan be re-enabled. Subsequent reads, re-reads, and writes can benefit greatly from DFC technology.