Thesis research: SSD vs. FC drive benchark tests – part I

typewriterI am writing my graduate thesis on the subject of Solid State Drive (SSD). By the way, D stands for DRIVE, not DISK, as SSD does not use disk. Now, with that out of the way…

I have been benchmarking a new SSD array that I have added to my companies SAN: an EMC CLARiiON Cx4-480 system running on 4Gb/s fiber. It will be 8Gb/s soon, but we are waiting on the NAS code (an EMC NS-40G) to catch up so it will support 8Gb/s: the firmware on the NAS only supports up to 4G currently. The SAN is held together with two Brocade 4900 FC switches.

About the disks that I will be testing and comparing:

Disks used: (5) EMC 70GB SSD and (5) 300GB FC disks.

SSD:

66.639 raw capacity -FC SSD – Manufacturer: STEC – Model: ZIV2A074 CLAR72 – Serial: STM0000E9CFD – 4Gbps

FC:

268.403 raw capacity – FC – Manufacturer: SEAGATE – Model: STE30065 CLAR300 – Serial: 3SJ09XWW – 4Gbps

  • Created RAID5 (RAID group 100) on five SSD model ZIV2A074, 66.639GB each.
  • Creating RAID5 (RG 101) on five 300GB FC disks: Seagate 15K.7 ST3300657FC
  • LUN 104 is assigned drive letter W: (disk 3) (RG100) and named “SSD
  • LUN 108 is assigned drive letter X: (disk 4) (RG101) and named “FC

The test server was installed and set up with one dual-port 4Gb/s HBA. Windows Server OS Standard with 1GB RAM.

SAN Management: EMC Navisphere 6.28.21.

Network: 4Gb/s fiber. 2-Brocade 4900B FC switches. Host HBA: Emulex LightPulse FC2243.

Host connection via EMC PowerPath v5.2

Test I/O is generated by Microsoft SQLIOSim; I/O generation utility to similate I/O patterns found within versions of Microsoft SQL server. The versions simulated are Microsoft SQL Server 2005, SQL Server 2000, and Server 7.0. Brent Ozar, a SQL expert, has a good video on using SQLIO on his web site at brentozar.com. I have learned some things from him and am using the tips on SQLIO for my benchmarking.

The monitoring will be done with EMC Navisphere Analyzer and SUN StorageTek Workload Analysis Tool (SWAT).

Here is a preliminary test on SSD vs. FC data rates using SQLIOSim to generate I/O and SWAT to record the results:

SSD:

SSD performance survey data rate

FC:

FC performance survey data rate

So far there is not much of a difference. The Fiber Channel drives are keeping up with the SSD. Of course, this is a preliminary test and other tests at this time are giving similar results. I continue to plan my testing methodology.

Advertisements

SSD

ssdFlash-based (solid state) storage – While the price is still a bit high for large storage pools, it makes complete sense for some applications (On-Storage, 2009). Also known as SSD (Solid State Drive), flash disks are touted to be the fastest storage medium yet – a great improvement to RAID technology and a performance boost to databases. According to EMC, solid state drives (SSD) can store and query data faster than the magnetic disk drive, including the modern fiber channel and SATA II disk drives, and are also more energy efficient. SSD’s can amass a terabyte of data using 38 percent less power than conventional FC disk drives. “Since it would take thirty 15,000 RPM FC disk drives to deliver the same performance as a single flash drive, this translates into a dramatic 98 percent reduction in power consumption to achieve similar transaction per – second performance” (EMC, 2008). Many have been waiting a long time for the rise of this exciting solid state technology. Eventually, this could mean the start of the end of magnetic media within the enterprise data storage environment.

Flash memory, also called flash RAM, is nonvolatile memory that is used by erasing and reprogramming memory blocks as needed by applications. It is a variant of electrically erasable programmable read-only memory (EEPROM). EEPROM is written at the slower byte level – flash RAM uses the more efficient block level. There are two types of SSDs: RAM-based and flash memory based. It is the flash memory that is changing today’s enterprise storage. Although flash is not as fast as RAM-based, flash is still the choice over magnetic media within the data center for fast data arrays.

Performance is superior to the legacy magnetic disk, and because of this advancement, there has evolved a new tier of storage: Tier 0. According to SearchStorage.com, “Tier 1 storage, also known as production storage, can be considered the first class cabin for production data. Tier 2 and lower storage tiers were developed to handle data that is not quite as critical or does not need the performance characteristics of Tier 1 storage (Searchstorage.com, 2009)”. Now with SSD, a new tier is defined for performance ahead of what tier 1 can offer. Previously, tier 0 was defined as RAM disk, and this was an expensive endeavor which demanded excessive amounts of RAM within systems. Now that the cost of SSD has dropped (and continues to drop), it is more accessible and more companies are adopting the new SSD tier 0 solutions for enhancing performance within critical information systems applications.

The benefits of SSD can be most realized within database stores, as these arrays demand a high level of I/O. The significantly reduced seek time on SSD can produce a high ROI over time. Another advantage of SSD is that it contains ECC (error correcting memory) technology, and there is no danger of file fragmentation which can reduce administration labor and service cost. This author has seen service cases in which a system was running slow due to severely fragmented file systems residing on magnetic SCSI disks.

It is a popular subject in information systems: are SSD’s green? Do they consume less power than the older spindle-based disk arrays? The answer is no – if one is looking at a power per TB comparison; but when looking at an array with a high spindle count to get greater I/O, the SSD can consume less energy. This is true because with a high spindle count (more disk drives) the more energy is consumed (especially within non-virtualized arrays); but with SSD, less drives are needed to get the desired I/O performance. According to SearchStorage, “SSDs do not need extra spindles; they deliver high speed out of the box. The result is a lower number of devices and therefore a lowering of power consumption rates” (SearchStorage, 2009). As for performance comparison, research has found that “A typical hard disk drive performs 4- to 5-msec reads or writes and approximately 150-300 random I/Os per second. A RAM-based SSD does .015 msec reads and writes and about 400,000 I/Os per second. A flash-based SSD does about 0.2 msec reads and 2-msec writes. I/O performance is 100,000 random I/O per second on reads and 25,000 I/Os per second on writes.” (SearchStorage, 2009).

SSD storage technology is contributing faster I/O and greater energy savings to enterprises. While it has not replaced the magnetic spindle disk, it will replace it in the not so distant future. This will benefit storage area networks of all types – from the small SAN to the global SAN by providing superior response times within database arrays and critical appliations.

References:

PUNT-IT. (2008). EMC Brings Flash Drives, Virtual Provisioning and 1TB SATA to SymmetrixRetrieved April 4, 2010 from http://www.emc.com/collateral/analyst-reports/pund-it-review-symm-flash.pdf

Koopman, J. (2008). Top 5 Storage Trends. OnStorage. Retrieved May 20, 2009 from http://www.on-storage.com/50226711/top_5_storage_trends.php

SearchStorage.com. (2009). Solid State Storage. Retrieved May 25, 2009 from http://searchstorage.target.com

King, C., IT, Pund., Hill, D. (2009). EMC Brings Flash Drives, Virtual Provisioning, and 1TB SATA to Symmetrix. Breaking News Review. Volume 4, Issue 4. EMC. Retrieved May 18, 2009 from
http://www.emc.com/collateral/analyst-reports/pund-it-review-symm-flash.pdf