This Storage Intelligence (StorInt™) briefing covers Microsoft Exchange 2010 and Exchange 2013/2016 Solution Review Program (ESRP) results . There have been no new 1000 and under mailbox ESRP submissions. However, there have been two new over 5000 mailbox submissions (Dell PowerEdge R740xd and Lenovo ThinkAgile HX7520 with AVH [Nutanix]) and three new 1000 to 5000 mail box submissions (Dell PowerVault MD3820F systems with FC SAN, FC Direct and SAS connected storage). Dispensing with SCI ESRP performance reporting tradition, we present new top ten charts from both the over 5000 and 1000 to 5000 mailbox submissions below.
Latest ESRP V3.0 & 4.0 (E2010&E2013) performance
In Figure 1 we show ESRP Top Ten Log Playback for the over 5000 mailboxes submissions.
In Figure 1, the new Lenovo HX7520 came in at 7th place with a 0.41 sec log playback time and was a hybrid (SSD-disk) storage solution Recall that log playback is measured as the time to playback 500 log files of 1MB each, under a standard database load. Log playback is a read sequential of the Exchange log while updating (applying the log to) the Exchange database, representing a mixed workload.
The Lenovo system was running Nutanix AVH across a cluster of 4 servers. It looks like the Exchange servers and the software defined storage were operating within the same server hardware as different VMs. This may have produced better performance as some log and database IO wouldn’t have needed to go outside the server. The Lenovo cluster had a total of 80 2TB 7.2Krpm disks and 16 1.92TB SSDs or 20 disk and 4 SSDs per node.
The Dell PowerEdge R740xd, disk only solution, didn’t place on any of our top 10 charts.
Next, we report on top 10 ESRP log play back for the 1000 to 5000 mailbox category in Figure 2.
In Figure 2, we can see the 2 out of the 3 Dell PowerVault 3820f solutions came in at 8th and (tied for) 9th place. That is the Dell PowerVault 3820f FC SAN and FC Direct was measured at 0.75 sec. and 0.81 sec. ESRP log playback, respectively.
In contrast to the Virtualized Lenovo solution above, Dell’s PowerVault Exchange configurations were more typical of standard deployments, in that they had separate Exchange (Dell PowerEdge R720) servers attached to standalone (PowerVault) storage systems. There was one Exchange server and storage array per site across two sites, with one site under test. Both the PowerVault MD3820f FC SAN and FC Direct submissions had 20 900GB 10Krpm disks.
Dell submitted a set of three PowerVault M3820fs with FC SAN, FC Direct and SAS storage connections. The SAS version came in at #11 with 0.86 sec log playback, placing it just outside the top 10.
In Figure 3 we show a chart depicting protocol performance that we typically reserve for our SAN NAS Buying Guide (available for purchase on our website, see below for more information).
In Figure 3 we can see the linear regression lines for FC (both SAN and Direct), iSCSI and SAS ESRP submissions. SAS shows up as the clear winner in database transfers per second per disk drive. According to the regression formula, for every disk in a SAS submission, it can perform ~61 database transfers/second. We also have this chart for log playback but SAS wins there as well and the correlations are better here. Similar charts were done for normalized database transfers per second and total database backups but the regression R2 were much worse there and as such, are not as dependable.
We truncated the plot at 50 drives as there’s a 100 drive FC solution but this doesn’t change the regression formulas. We also limit submissions on this chart to be disk only submissions with 10,000 and under mailboxes that used 10Krpm disk devices.
One problem with this chart is the relatively few SAS and iSCSI submissions, at five each vs. FC which has a total of 18 (one of which is hidden). Fewer submissions can impact the linear regression accuracy, which shows up as lower R2.
Unfortunately, the Dell PowerEdge R740xd didn’t rate on any of our top ten charts. There still have been no new under 1000 mailbox submissions now going on over 4 quarters. But, as we have broken with SCI ESRP reporting tradition and reported on new top ten charts for all ESRP categories, that may not be a problem anymore.
Constructive comments on how to improve our analyses for Microsoft ESRP (Exchange) or any of our performance reports are always welcome. Moreover, if you detect errors in this or any of our other performance reports, please do let us know and we will correct them as soon as possible.
If you received this report from someone else, please consider signing up for your own free copy of SCI’s Storage Intelligence (StorInt™) monthly e-newsletter using the QR code below right. We provide commentary on recent major storage system announcements as well as analyze current storage system performance results/benchmarks every month. We plan to analyze Microsoft ESRP results again in three months.
[This performance dispatch was originally sent out to our newsletter subscribers in January of 2018. If you would like to receive this information via email please consider signing up for our free monthly newsletter (see subscription request, above right) and we will send our current issue along with download instructions for this and other reports. Dispatches are posted to our website at least a quarter or more after they are sent to our subscribers, so if you are interested in current results please consider signing up for our newsletter.]
[Also we offer more block storage performance information plus our OLTP, Email and Throughput ChampionsCharts™ in our recently updated (May 2019) SAN Storage Buying Guide, or for more information on protocol performance results please see our recently updated (May 2019) SAN-NAS Storage Buying Guide, both of which are available for purchase on our website.]
Silverton Consulting, Inc., is a U.S.-based Storage, Strategy & Systems consulting firm offering products and services to the data storage community