San Testing round 1

Decided it was time to do some testing on my sans to get an idea of what iops they are capable off. I used a windows 2008 r2 base server to start my testing, with the MS mpio drivers. Test were ran using the sqlio tool from microsoft. I will add the vendor mpio drivers later. I am testing the EMC VNX, NetApp FAS3210 , IBM DS8100 behind IBM SVC and VNX behind IBM SVC. SVC and the VNX are on the 8Gbs FC, while the FAS3210 and IBM DS8100 are on 4gbs FC.

The VNX is using auto tier (FAST VP), wtih 200 GB of FAST Cache, the Raid 5 storage pool that has 100 gb SSD, 600gb and 300gb 15K, 300gb 10K and 2tb 7.2K drives in it.

Netapp FAS3210 I am using an Aggregate with 15 1tb 7.2 k disk.

DS8100 is a Raid 10 with 96 73gb 15k disks.

The first round of test were done with w2k8 r2 right out of the box using the windows native mpio driver. SQLio is using a 64k block size and writing or reading a 50gb file. Here are the Results. As expected the 8100 was the best performer, the NetApp and the VNX were Close, in fact the Netapp beat out the VNX .

Storage System iOs/Sec Mbs/Sec Ran/SEQ io Read or Write
8100 4700.85 293.8 Random Write
8100 4820.72 301.29 Sequential Write
8100 15115.02 944.68 Random Read
8100 13950.89 871.93 Sequential Read
VNX/Auto Tier 445.94 27.87 Random Write
VNX/Auto Tier 3783.15 236.44 Sequential Write
VNX/Auto Tier 4007.63 250.47 Sequential Read
VNX/Auto Tier 1570.34 98.14 Random Read
NETAPP 883.45 55.21 Random Read
NETAPP 284.96 17.81 Random Write
NETAPP 4503.39 281.46 Sequential Read
NETAPP 2725.82 170.36 Sequential Write
VNX/SVC 4638.99 289.93 Sequential Write

The next set of test I loaded SP1 for w2k8, I know there were some improvements in the mpio driver, however… I was not expecting the results I got below.  The VNX and the 8100 made huge improvements on write where reads were a little slower with the new mpio driver. However the FAS3210 had a decrease in io across the board. Later on I will load the Vendor MPIO drivers and see what the results are.

Storage System iOs/Sec Mbs/Sec Ran/SEQ io Read or Write
8100 7327.36 457.96 Random Write
8100 7439.3 464.95 Sequential Write
8100 10815.27 675.95 Random Read
8100 16070.83 1000.42 Sequential Read
VNX/Auto Tier 7372.79 460.79 Random Write
VNX/Auto Tier 7168.43 448.02 Sequential Write
VNX/Auto Tier 3001.21 187.57 Sequential Read
VNX/Auto Tier 6668.09 416.75 Random Read
NETAPP 681.34 42.58 Random Read
NETAPP 693 43.31 Random Write
NETAPP 2147.03 134.18 Sequential Read
NETAPP 1378.91 86.18 Sequential Write
VNX/SVC 4593.66 287.1 Sequential Write
VNX/SVC 632.38 39.52 random Write

More to come later I am going to tweak the block size and see if I can find were they perform the best.

XtremIO

Had a chance to go to an EMC XtremIo briefing this week. XtremIO is EMC’ss entry into the all flash array vertical. The presenter made and interesting point right off the start, which was that it as not about the flash in the array, but about the software. XtremIO takes all the data coming in and hash’s it into 4k bits, and the writes data to flash when it has a full stripe. It also does in-line de-duplication. The performance numbers for one X brick are 150,000 iops with random reads and writes on an array that is 50% full. These numbers were also given when the array is getting data that is greater than 4k blocks, so these appear to be real world numbers. Performance will go up (essentially double with each x brick) as you add X bricks, right now the max  will be 8 X bricks, but the software has been designed where there is no limit on the amount of x bricks that could be in a XtremIO flash array. The files system is not a traditional raid it is n+2 and was built specifically to handle an all flash array. Native replication for this array will be out next year and it will only offer asynchronous replication , as synchronous would affect the overall performance of the array. If you need replication now you can use vplex, and native support for Recover point will be coming and the end of this year.

I like the scalability, the fact you can use all off the array (some vendors recommend not to use more than 80% capacity on their flash array’s). They only other flash array I can compare this too is the IBM flash systems with IBM seems to recommend deploying behind SVC , which I understand since the FlashSystems are not scalable and do not have any replication natively.  With XtremIO you will not need a second device for replication and if you need more space you can just add another x brick and get more performance at the same time. XremeIO looks very interesting to me and it will be interesting to see how it is adopted in the market place and what other Flash Arrays are coming from other vendors.

 

IBM Edge Day 4

This will probably be my last post about Edge I only have time for one session tomorrow.
First session was a session about the future of IBM’s storage virtualization products (SVC/Storwise). Coming features include three tiers for easy tier new 2u hardware that has the ups built in with configurable processors and cache, multiple FC/fcoe cards and eventually encryption. The spoke more on the change that allows system i connect to SVC/storwise and it appears the change was made on the system i side not storage. Meaning the currently releases version of SVC code 6.4.1.4 can connect to system i assuming you have the correct code level on the i. Since the change was made on the i, I do wonder if that means finally any storage from any vendor can now connect to the system i.
I went to a v7000 unified class the presenter spent most of the time open how to get the file side setup. Unfortunately this reminded me of a typical IBM experience in the past it seemed overly complicated and a lot of command line configuration. He went over setting up ip’s and permissions in my experience this is way easier to do on both Netapp and EMC file offerings.
Over all I have enjoyed IBM Edge. Before I came I am not sure i would have considered a storwise or a pure system. If you asked me know to choose between a flexpod, vblock, or pure flex I would probably choose the pure flex , mainly because of the storwise storage that would let me virtualized any other storage I might have. I think Netapp needs to get on their game from talking to other customers Netapp’s sans are much more expensive than both EMC VNX and IBM v7000 and I believe that both EMC and IBM will continue to take market share from Netapp in mid tier storage market.
I only had a few complaints about IBM Edge. The first one is the quality of the tweets about edge were very low,mostly marking speak did not see a lot of discussion or opinions, this is not a huge deal at but the next one is. Next year more computers for labs and a better lab setup. I attempted to go to 4 labs and was unable to do any of them. There seemed to be 12-16 computers for the labs and with 4000+ people that is not going to be enough. However even worse were the computers could not do every lab and some case I saw 4 people grouped together to work on one computer. IBM Edge has told me via twitter that the will have more computers next year.
IBM seems to be doing what other “Big Data” companies have been doing, they have introduce things to make IT admins life easier such as a GUI management system for TSM. The are unifying the look of their management GUIs so they feel the same no matter what IBM storage product you are on. They may have been late to the flash and unified (pure flex) system game but they are all in with what appear to be very good products.
The only real question I have for IBM’s future will be what happens with their server division if it is sold off and how that will affect the power and flex systems.

IBM Edge day 3

So no more general sessions just presentations for the rest if the week.
I finally had that wow moment today. On June 1st IBM announced native support for direct/fabric attach of IBM i to v7000/SVC storwise. This is huge for us as we were looking to replace our 8100, 4800, 6800, 5100 and VNX with a new Enterprise storage system and consolidate on one platform. However due to budget constraints it was looking like we were just going to have to keep the VNX and 5100 and use our 8100 our colo facility instead of the VNX. We still need to work out the logistics however in my head 2 v7000 at our main site and take our 4 node Svc cluster to colo with the VNX. This could save us money in lots of way the only obstacle i see right now is we were looking to do encryption on the 8870, so we may need to look into some Brocade encryption switches.
IBM new blade system which they call a gen 2 blade system is called flex system,a coupe if things stood out here to me. You can order power i blades , actually any power blades and all inter blade server communication stays in the chassis. I do not know if you can manage and configure multiple blade chassis and servers in one place like you can with Cisco UCS, from what I was able together you cannot however the flex system can hold 14 blades vs the UCS 8.

IBM Edge day 2

Day 2 started of with more of Delta Rae in the general session which was an excellent way to start out the day.
There was more talk of flash arrays, Tom Rosamilia brought up a person from sprint. They have implemented flash arrays and seen a 45x improvement in response time in their applications and are seeing cost savings due to flash arrays taking less power and less space. As a mid tier customer I do wonder what the ROI is on an all flah array if you do not need them from performance due to power and cooling reduction.
Dr. Bernie Meyerson came up and talked about cyber threats and how using analytics can help you be agile enough to respond to unknown attacks before loss of data happens. He had a chart that showed that the number of unknown attacks I on the rise. Unknown attacks as in attacks were data was stolen and only realized after the fact, and worse no one knows how it happened.
Next was a talk by Dr. Samuel Nussbaum about how Watson was changing healthcare and how it was being used to pick the best treatment plan for oncology patients. While it was very cool to see how Watson is changing health care I am not sure from a technical point of view how valuable this was for me.
Last was David Becker from Pew charitable trust and Jeff Jonas from IBM to talk about how his G2 engine help 7 states increase voter registration and figure out if people had moved or not. I would like to have heard more fromJeff Jonas he was an excellent speaker.
Went into some more technical sessions overall had better success with the sessions I chose ironically I only had 2 different presenters Jim blue talking about Fiber channel technologies and more with Brian Sherman about DS8000 mirroring and Svc. Jim and Brian are both excellent presenters and very knowledgeable.

Day 2 ended without me yet having a real wow moment from learning/seeing something I need to get at work asap, hopefully tommorow.

First post from IBM Edge

Hi from IBM edge this is the first post on my new storage blog. I arrived at Edge on Sunday and the first general session was Monday morning. I enjoyed Charles Long CEO and founder of Centerline Digital, he mention how their pure system implementation was letting them do things 4X faster, I would have liked some more info on what the Pure System replaced to make it 4X faster. Centerline Digital http://centerline.net/ did all of the onscreen graphics at Edge. IBM is now pushing all flash arrays as a next game changer in storage. Late last year IBM bought Texas Memory Systems makes of the RamSan devices an all flash array. This is different than SSD disks as this is a chip array and I believe they were throwing around 850,000 iops guarantee as a number and 100 microseconds of latency, Ambuj Goyal spoke about how storage is now a Forethought for IBM and mentioned the 6 Billion Dollars in R&D IBM is spending (1B just for flash) to make storage and associated technologies a forethought. What stuck with me about this statement was IBM’s Main competitors in the storage area storage has always been a forethought since that is the main part of their business. I do think that with Pure Systems and Flash System IBM does seem to have a real roadmap that to me seemed to be lacking in years past. They also pushing the Software Defined Storage and doing it “right” what with Open api’s and integration to open stack, they also took as small shot at EMC by saying that Viper is the new player and they are into thier 3rd generation SDS with SVC/Storwise. I had never thought of SVC as SDS platform and I can buy that however if you are going to consider SVC as an SDS platform it seem vPlex would be a more apt comparison and viper could be considered EMC’s next generation SDS platform. IBM seemed to mention open quite a bit mostly in relation to open stack. I do find it ironic that a few years if you looked up proprietary in the dictionary you might have seen a IBM logo. However I do think is a good sign as I do see a lot of positive change in the way IBM does business with the announcements in the Day 1 General Session. I do wonder about the future of Pure Systems if the rumored sale of IBM server division was to happen.

As far as the tech sessions I was unable to attend my first sessions as there was no room in the inn, the second session I went to was a Lab with about 18 people in however, they only have 4 computers set up to do the lab for Easy Tier and Storwise. So my morning was a Wash, I went to some second choice tech sessions one including one about Tivioli Storage Manager new features, IBM has finally released a GUI management console for TSM and all I can say is, ITS ABOUT TIME. My afternoon sessions were much better Brian Sherman did 2 of them about the DS8870 and SVC/Easy Tier and they were both excellent. IBM has new microcode update for the 8870 7.1 which include VAAI for vmware, however these features are not available for any other DS8000 series, several people gave feedback about this to them and hopefully they will release these features for these older systems as some are quite viable (DS8700, DS8800). Forgot to mention the Delta Rae played at the general session and they were excellent. I high recommend you check them out. http://www.deltarae.com/

Jim Gaffigan was the evening entertainment and he was excellent.