Sign in with
Sign up | Sign in

LSI MegaRAID 9265-8i

Four SAS 6 Gb/s RAID Controllers, Benchmarked And Reviewed
By , Achim Roos

According to LSI, SMEs (small and medium enterprises) are the target audience for its MegaRAID 9265-8i. The company markets the card as well-suited for cloud, security, and business applications. With a street price of approximately $630, the MegaRAID 9265-8i is the most expensive controller in this test, but as the benchmark results show, you get what you pay for. Before we present the benchmark results, let’s discuss this controller's technical features and its optional software add-ons called FastPath and CacheCade.

The LSI MegaRAID 9265-8i is based on a dual-core LSI SAS2208 ROC, employing an eight-lane PCIe 2.0 interface. The suffix -8i in the product name denotes eight internal SATA/SAS ports, each of which supports 6 Gb/s. Up to 128 physical storage devices can be connected to the controller via SAS expanders. The low-profile card also features 1 GB DDR3-1333 cache, and supports RAID levels 0, 1, 5, 6, 10, and 60.

Tuning Tools, FastPath, and CacheCade

LSI claims that FastPath can dramatically accelerate the I/O operations of attached SSDs. The company says the FastPath feature works with any flash-based SSD, markedly improving the read/write performance of a SSD-based RAID array by up to 2.5x in writes and up to 2x in reads, achieving 465 000 IOPS. We weren't even able to put that to the test, though. As delivered, this card already had enough horsepower to handle our five-drive SSD array without FastPath installed.

The other software option for the MegaRAID 9265-8i is called CacheCade. It enables the use of one SSD as a read cache for an array of hard disks. According to LSI, this may accelerate read operations by a factor of up to 50, depending on the size of the data being accessed, the application, and the use case. We tried this tool out and created a RAID 5 array consisting of seven hard disks and one SSD (the SSD serving as a read cache). Compared to a RAID 5 setup composed of eight hard disks, it's evident that CacheCade not only improves measured I/O throughput, but even perceived performance (increasingly so as the original data set becomes smaller). We used a 25 GB data set for our test and achieved 3877 IOPS in the Web server Iometer workload, whereas a normal hard disk-based array was only able to hit 894 IOPS.

Performance

In a nutshell, LSI's MegaRAID 9265-8i is the fastest SAS RAID controller in this round-up with respect to I/O performance. As far as sequential read/write operations go, however, the controller only achieves mid-range performance, as its sequential performance varies greatly depending on the RAID level you use. In the RAID 0 hard disk test, we see sequential reads of up to 1080 MB/s (significantly faster than the competitors). Sequential write performance in RAID 0, 927 MB/s, takes a first-place finish, too. In RAID 5 and 6, however, the LSI controller is beaten by all of the other contenders, only redeeming itself in the RAID 10 benchmarks. In the SSD RAID test, LSI's MegaRAID 9265-8i posts the best sequential write performance (752 MB/s) and is only beaten by the Areca ARC-1880i in sequential reads.

If you are looking for an SSD-oriented RAID controller with high I/O performance, the LSI controller is a winner. With rare exceptions, it winds up in first place in our Iometer database, file server, Web server, and workstation workloads. When your RAID array consists of SSDs, LSI's contender is truly unleashed. It utterly outclasses the three other controllers. For example, in the workstation benchmark, the MegaRAID 9265-8i achieves 70 172 IOPS, while the second-place finisher, Areca's ARC-1880i, posts slightly more than half that number, 36 975 IOPS.

Display all 46 comments.
This thread is closed for comments
  • 1 Hide
    americanherosandwich , October 6, 2011 4:45 AM
    Great review! Though I would have like to see some RAID 1 and RAID 10 benchmarks. Don't usually see RAID 0 for expensive SAS RAID Controllers, and more RAID 10 configurations than RAID 5.
  • 1 Hide
    purrcatian , October 6, 2011 7:14 AM
    I just sold my HighPoint RocketRAID 2720 because of the terrible drivers. Not only do the drivers add about 60 seconds to the Windows boot, they also cause random BSODs. The support was a joke, and the driver that came on the disc caused the Windows 7 x64 setup to instantly BSOD even though the box had a Windows 7 compatible logo on it. I even RMAed the card and the new one was exactly the same.
  • 0 Hide
    dealcorn , October 6, 2011 12:08 PM
    Very cool, fast and expensive means not home server stuff. For that, try the IBM BR10I, 8port PCI-e SAS/SATA RAID controller, which is generally available on eBay for $40 with no bracket (I live for danger). You are stuck with 3 GB/sec per port, but if you add $34 for a pair of forward breakout cables you have 8 sata ports at a cost of under $10 per port. The card requires a PCIe X8 slot but if you only give it 4 lanes (the number of lanes offered by our Atom's NM10) if will give each port 1.5 Gb/sec. Cheap SAS makes software RAID 6 prudent in a home storage server.
  • 0 Hide
    slicedtoad , October 6, 2011 1:29 PM
    I have pretty much no use for anything other than raid 0 but it was still an interesting read. I think i prefer this type of article over the longer type with actual benchmarks thrown in (not for gpu or cpu reviews though).
  • 2 Hide
    rebel1280 , October 6, 2011 1:49 PM
    Great read! Way better than rumors and junk, stick with this kind of stuff Toms!
  • 0 Hide
    pxl9190 , October 6, 2011 4:17 PM
    Only wish this review had came earlier !

    I had a hard time deciding between 9265-8i, 1880 and 6805 a month ago. I bought the 6805 and always wondered why RAID-10 was not as fast as I thought it should be. This reviewed proved my worries.

    I eventually went to RAID 6 with 6 Constellation ES 1TB disks. Here's where the adaptec really shines. This is for a photo/video storage/editing disk array.

    Admittedly if I have a choice again I would have picked the Areca after seeing the numbers. Adaptec was the cheapest among all of them so it's not too much of a regret.
  • 0 Hide
    Anonymous , October 6, 2011 4:47 PM
    Great review! As I am in the process of building a new home file server and always have a habit of going overboard in such situations, I will be referring back to this article many more times before purchasing.

    That said can you please talk more to the differences performance wise between SATA and SAS? I understand the reliability argument, however I wonder if for my purposes I would not be better served by using cheaper SATA disks over SAS disks?

    I would also love some direction with regard to a good enclosures/power supplies for a hard drive only enclosure. I realize I am quickly priced out of an enterprise solution in this arena, but have seen at least a couple cheaper options online such as the Sans Digital TR8M+B. (This enclosure is normally bundeled with some RocketRaid controller which I would probably discard in favor of either the Adaptec or LSI solution.)
  • 0 Hide
    Anonymous , October 6, 2011 5:07 PM
    You are missing a huge competitor in this space. Atto RAID Adapters are on par and I think the only other one out there, why are they not compared in this review?
  • 0 Hide
    Marco925 , October 6, 2011 5:18 PM
    I bought the Highpoint, for it's money, it was incredible value at a little under $120
  • 0 Hide
    stuckintexas , October 6, 2011 6:33 PM
    I evaluated all but the Highpoint for work. What isn't shown, and would be unrealistic for a home user, is that the LSI destroys the competition when you throw on a SAS expander. With 24 15k SAS drives, the LSI card tops out at 3500MB/s, RAID0 sequential write, while the Areca is
  • 0 Hide
    stuckintexas , October 6, 2011 6:36 PM
    Sorry for the double post, comment system doesn't like the less than character.

    I evaluated all but the Highpoint for work. What isn't shown, and would be unrealistic for a home user, is that the LSI destroys the competition when you throw on a SAS expander. With 24 15k SAS drives, the LSI card tops out at 3500MB/s, RAID0 sequential write, while the Areca is less than 2500 and the Adaptec is less than 1800. The Areca also has a lot of issues with stuttering during writes, your average may be fine, but the throughput has some significant dropouts.
  • 0 Hide
    fenwickc , October 6, 2011 6:56 PM
    How do these cards compare with using the 6x SATA 2 connections on my motherboard, a couple of cheap $30 2 port SATA card (eg StarTEch PEXSAT32 2-Port PCI Express SATA 6 Gbps) and software RAID 6?

    I have more CPU that I can use (core i5) and want to use cheap 2 or 3 TB 7200 rpm SATA drives because I want lots of storage rather than maximum speed.
  • 0 Hide
    marraco , October 6, 2011 7:26 PM
    When we will have asymmetric RAID 0? A RAID controller capable of splitting data on different sizes parts, so the largest parts go to the fastest drives, and the shorter ones to the slower drives.
  • 0 Hide
    jandabaer , October 6, 2011 8:32 PM
    I'd also love to see a comparison between these controllers and software-RAID on the Intel Sandy-Bridge-E platform so see if we can believe Intel's marketing:

    http://intelstudios.edgesuite.net/idf/2010/sf/aep/STOS002/STOS002.html
  • 0 Hide
    Anonymous , October 6, 2011 9:10 PM
    Great review, however I would have liked to see more details around raid configuration for each card. Things such as:
    1. Supported raid features
    2. Raid rebuild rates, notification features, etc
    3. Gatcha's with each card, ie are JBOD disk interchangeable between different raid cards.

    I am not surprise to see the HighPoint's card at the bottom of the list. You really get what you pay for with these cards, poor performance and even poorer support. I have a RocketRaid 2320 which has horrible drivers and sucks in every category. Will never use another HighPoint card due to the mounting issues I have encountered.
  • 0 Hide
    palladin9479 , October 7, 2011 2:09 AM
    A slight correct to the Article about FC controllers. You don't use FC for raw speed you use it for redundancy and multi-pathing. A single FC drive will connect to two different channels, each channel can go back to the same HBA on two different channels or to two completely separate HBAs. This way each drive has at least two channels to reach the host system. Also FC comes in 2, 4, 8 and 10 Gbps flavors, kinda crush's SAS-6 in raw bandwidth. Although honestly you won't see faster then 4 or 8 on the inside of a system, 10 is usually reserved for between SAN drive arrays and SAN fabric switchs. With multi-pathing not only are you getting redundant connections, you can mux the two path's to combine their bandwidth. A system sporting two dual 8Gbps HBAs would be communicating to the SAN at 32Gbps across four connections to two different switches.

    Which brings up the last point, FC's expandability is beyond SAS and FIS PM/PE's. PM/PE was designed for BBC connections where you have a single channel to a back plane with four to eight hot swap SAS connectors. And while they left room for you to implement 255 ID's per channel, there isn't a single vendor who provides that solution. FC on the other hand is as expandable as Ethernet. you can just keep adding more drive arrays, as many as you want. Each storage processor has it's own limit, usually around 255 disks, but you can just add more storage processors.

    That all being said, FC is for enterprise class storage networks. Its the absolute best protocol for that due to its expandability and scalability (disks + bandwidth). SAS is for local system disks on small to medium business servers. Any enterprise worth it's salt will be using VM technology with the VM's being stored on the SAN for availability / redundancy purposes.
  • 0 Hide
    mras , October 7, 2011 9:01 AM
    "Aside from their performance characteristics, they stand apart by offering handy features like mixed-environment SAS and SATA support, along with scalability via SAS expanders."
    Can't you test those statements in a upcoming article?
    My personal experiences says that the HP Sas expander, works flawless with the LSI and Acera card you tested, with both single and dual linking.
    However, the Adaptec only seems to understand single linking, while the Highpoint doesnt work at all with it.
  • 0 Hide
    g00ey , October 7, 2011 11:21 AM
    With my experience of losing a lot of data due to failing hard drives one motive to build a storage cluster on a dedicated controller is reliability.

    I myself have built a storage pool using ZFS operating the SAS controller in IT-mode (Initiator-Target mode which means that all RAID functionality is turned off which it should be when using ZFS). So you don't buy an expensive hardware RAID card for that, instead you buy a cheaper card with lower RAID functionality. The RAID is instead taken care of by the software which has shown to be a lot more reliable than hardware RAID solutions. The IT-people at CERN who process petabytes of data every day can testify to that when they operated a huge storage cluster built on Areca cards; In short, the hardware RAID wasn't as reliable as promised whereas the ZFS software RAID solution was.

    When using an operating system such as Solaris or OpenIndiana, one really important property of the controller is the platform compatibility. There are currently only two brands that can hold up to compatibility and that is LSI and Intel. LSI are known to be especially reliable and most thoroughly tested as most operating system vendors provide native drivers for use of LSI hardware in server environments and they have been used in such environments for years by now.

    Brands such as 3Ware and OEMS such as Dell, IBM, Intel, HP, Fujitsu-Siemens, Cisco et al build SAS cards that are mostly based on LSI chips (look for MegaRAID 1068e/1078e or 2008e/2108e chips in the specs).

    From a compatibility standpoint the Highpoint cards is the last brand I would recommend and from a reliability standpoint I would certainly recommend people to stay away from anything that comes from JMicron.
  • 0 Hide
    scrianinoff , October 7, 2011 7:16 PM
    Interesting review! What I find really disturbing is that results obtained by others using the Highpoint 2720 are much better while being consistent with each other, such as here:

    and here:

    Are the others lying, have they done it all wrong, or was there something wrong with Tomshardware's setup or drivers?
  • 0 Hide
    scrianinoff , October 7, 2011 7:19 PM
    Now with working links, sorry for the double post:

    Interesting review! What I find really disturbing is that results obtained by others using the Highpoint 2720 are much better while being consistent with each other, such as here:

    and here:

    Are the others lying, have they done it all wrong, or was there something wrong with Tomshardware's setup or drivers?
Display more comments