Future Storage Systems: A pause in workflow

Since I started this article series, I’ve had the awesome opportunity to have my ideas (well, some of the early articles at least) reviewed by person(s) who deal with the actual infrastructure of storage systems day in, day out.  The benefit of such peer review is that you get to learn at the symbolic “feet” of the masters and discover flaws, omissions, and understated features that need to be understood and incorporated.  This post is dedicated to some of those discussions and, where applicable, my understanding of how the FSS either incorporates or misses the boat.

Future Storage Systems: Part 4: Operating System – Conceptual Overview

In the previous Future Storage System articles, we’ve covered the basic hardware foundation for what I envision to be a powerful future-oriented storage solution for the commercial midrange.  However, as you’re probably aware, hardware is meaningless without software to provide the operational capabilities that are needed to mange information.  In this article, I will focus on a general design for an extensible software layer (an OS) that will provide future-oriented capability expansion as well as robust analytics, capabilities, and integration with business continuity principles.  As always, please reference the diagram below.

Future Storage System - Operating System - Conceptual

Future Storage Systems: Part 3b – I/O Expansion Node

In Part 3a, we discussed the possibility of a purpose-driven Compute Node based on the Torrenza initiative for the Future Storage system.  This expansion node made use of Hypertransport as a “glue” between the base storage compute node and the expansion node (of computation or I/O flavours) that could be added.  The advantages of that topology were simple:  hot add support for additional processing power, additional I/O bandwidth within the system, and additional computing power for the array OS (which we’ll cover in a later article).  In this overview, we’ll take a look at another variation on an expansion node: an I/O expansion node that will add additional front-end ports and/or functionality to the base system.  We will be referencing the diagram below. (Apologies in advance for the image shearing off in the lower right hand corner).

Hypertransport I/O Expansion Topology

Future Storage Systems: Part 3a – Node Expansion Overview

In the previous two articles on the Future Storage System (FSS), I took a general look at a basic storage system architecture (Part 1) and then went a bit deeper into some of the more interesting bits of that system from a platform standpoint (Part 2).  In this article, I want to dive a bit deeper into how I envision nodes to be building blocks for additional capabilities and processing directives.  I will be referencing the image below as part of this article.

Hypertransport Node Expansion (detailed)

Future Storage Systems: Part 2 – Detailed Node View

So, in my article yesterday, I gave a global view of a very simple storage system for the future. Since I LOVE this type of conjecture and theoretics (is that a word?), I decided to take this a step further and flesh out some of the other intricacies of the design.  Check out the image below and then click through to read the rest.

Fleshing out the Hypertransport Storage System

Snoop Filters and the Tyan S5397

Just to add to the ever-growing information data base on the excellent Tyan S5397.

If you seem to be having problems with reported memory bandwidth in your favourite worthless synthetic benchmark or reality-based simulation workload, try disabling the Snoop Filter in the bios.

Remember, the Intel Xeon processors use a different set of prefetch schedulers than standard “consumer” level parts, so, this can also affect workload.

Search Term Tuesday – June 3rd Edition

Note: I’m trying to tighten up the layout of content on the landing page, so, I’ll be using excerpts more and more.

Continuing from last week, this edition of Search Term Tuesday will tackle the most important searches (or at least highest ranked statistically) that landed you at this blog.

Search Term #1: cx3-10 create raid group

I *heart* SMB

I Love SMB
There’s somewhat of a misconception that EMC dislikes the SMB market. While it’s true that we’ve largely ignored it, our partners in the channel (Dell, CDW, to name a few) have successfully brought our Clariion lines to the mid-level companies out there and have solved their storage needs through aggressive cost-cutting, etc. All that was well and good, but, still, there was a general sense that we were ignoring the small guys out there who had explosive growth potential for storage and revenue and who wouldn’t be able to buy in to commercial grade arrays. Enter the Clariion AX4-5 line.

In my current role, I work almost exclusively with customers who are either new to EMC or who have admired from afar our capabilities for the commercial market. They’ve perhaps been exposed to EMC through co-branding and software bundling in external hard drive technologies, smaller ASN partners that are localized by geography, or, through a channel source like CDW. Truthfully, the greatest “capture” of these customers has been through the channel and partners who have spent the time on developing relationships with their customers versus the standard “Hey, buy my product…” One of the greatest challenges then, is for EMC to step in to the ring and actually become more “friendly” or approachable with our product and personnel.

I have been critical of the Iomega acquisition in the past based on what I felt were misappropriated metrics (namely, “buying” our way into the SMB channel using stale products versus innovative new products). The Iomega name, which was once an entity noted for innovation, has diluted to nothing notable in the channel. Companies like Enhance Technology, Buffalo, etc. which continue to innovate in the SMB space have experienced great growth because they’ve continued to innovate and make their technology approachable to the masses. That being said, here’s what I see EMC doing in the SMB space.

Intel or Iomega?
It’s no secret that EMC has a special place in their heart for Intel. Our arrays (with the exception of the Symmetrix) are built around their processors, we are firmly esconced in the Win-Tel alliance with Microsoft and, well, you know the rest. We started our SMB push with rebranding Intel 4200-E units and coupled them with EMC’s Lifeline (or is it Fortress?) software. While the 4200-E is a great box, I personally viewed the “Baxter Creek” 4000-E with a little more favour since it allowed for NFS, CIFS within the same box. The 4200-E has media functions, disk protection, etc. but it is truly the software that makes the box anything special. As I’m no expert regarding the software, I’ll defer to that product team for guidance. In any case, with the Iomega acquisition, Intel will need to be relegated to a smaller portion of the EMC market to allow for Iomega to gain some traction with SMB.

Iomega brings a much needed “middle ground” that the Intel boxes cannot cover. You’re looking at budgetary allocations between $1,000.00 to the top end of around $15,000.00 based on features and pricing. Now, I don’t know exactly what they’ll be changing or re-aligning (and one has to hope they’re going to stop this ridiculous whitebox rebranding stuff they’re doing now….) but chances are we’ll see DAS and NAS offerings that will allow for SMBs to start consolidating their storage. At risk is virtualization support and integration, but honestly, most of the offerings in that market have a buy-in around $6,000.00+ (based on the product, no less), so, it could be considered part of a larger spend for that company. In any case, I’d expect EMC branded software (Lifeline?) to appear in the boxes with some level of larger integration with our Core offerings. *shrug*

Iomega or EMC Core?
In the notes above, I’ve laid out how Iomega and Intel will be at odds with each other within the same market space (very briefly). In this section, I’d like to examine potential intersection points with Iomega and EMC Core products (Clariion, Celerra).

The Clariion AX4-5 units have been positioned as near-line SMB/Commercial storage arrays. By near-line, I’m simply referring to their ability to straddle both SMB and Commercial, not necessarily their data storage/retrieval capabilities (though, to be honest, the AX4-5 is a GREAT box to have). With our partners, you can get ready-made AX4-5 units starting around $9,000.00 (minimum config, to be honest) and scale into full featured SAS/SATA powered SAN storage for (obviously) more than that. A lot of the configs that I’ve worked on have hit the sweet spot for AX4-5 entry (with typical app integration with VMWare and SQL, for example) hitting around $35,000.00 all in. That’s not a bad place to start, especially since you get enterprise grade SAS and SATA drives that we’ve tested the living stuffing out of. Same applies for the EMC Celerra NS-2x line. I’ve worked with configs that will slap high IOP fibre drives and long term storage SATA drives within the same array for around 15% more than the AX4-5s. Again, this is optimized by the environments that I’m scoping out and, in most cases, this represents the best fit for performance and capacity. How is Iomega going to fair in that world? I’m not sure.

Nothing that Iomega has put forth so far really strikes anyone with any sort of wow-factor or “I must have it NOW.” For example, the high end 450R is simple a 4 drive server that has two GigE ports on the backend. For small shops, this may be worth it, but, again, the Intel “Baxter Creek” 4000-E does the same thing for about $4,000.00 less, fully configured. Now, you can’t really compare it against an AX4-5, to be sure, but, even a single SP AX4-5 is going to start you out around $4,000.00 above the 450R with better management. Put this into a financing or leasing model and….well, you’ll make out like a bandit. Truth be told, I’m assured that new products with new innovation will be coming down the road so, I’ll reserve my final judgment for that period of time, however, the clock is ticking.

Time to innovate…
EMC has the resources and ability to create new and exciting product at a price any SMB can afford. What they cannot afford to do is miss this opportunity. They’re counting on Iomega’s reputation and channel presence more than anything to drive their business home to the millions of SMB customers out there. The problem is, reputation and presence only go so far. There has to be better products brought to the table or else, competition will undercut and drive us out. We can’t rest on our laurels as a storage entity (truth be told, EMC isn’t well known in the SMB space at all…) because SMB doesn’t care. What matters the most is that they feel protected in their growth, from a storage AND business perspective. A misplaced investment on yesterday’s technology can destroy the confidence and ability to grow of ANY business, but is particularly pertinent in SMB.

SMB drives innovation; it breathes it out with every swell of its ranks. It is the cornerstone of tomorrow’s Commercial accounts and if we screw it up with SMB now, we’re going to have a hard time taking that ill taste from their mouths at the commercial level.

cheers,

Dave

Seed Newsvine

Piggy-back concepts of “Greening” the datacenter

I’ve had a LOT of fun, lately, reading Mark Lewis’ blog (found here) as he delves into the green data center concepts. To rehash some of what has already been talked about to “green” your data center:

a.) Tier your storage. Higher speed spindles, by nature, consume more power. (Compare the specs for the Seagate Barracuda ES.2 Enterprise SATA drive to those of the Seagate Cheetah 15K.5 FC/SAS drives). By moving your data from higher speed spindles to lower speed spindles based on usage/access patterns within a larger system policy framework, you can keep power consumption low overall. Better yet, archive it off to a Centera and remove the need for tiering within the array to begin with.
b.) Virtualize, Virtualize, Virtualize. Sure, it’s the “trendy” thing to do these days but, with the ability to collapse 30:1 (physical to virtual) in some cases, simply investing in VMWare (of course) will cut down on your power footprint and requirements. From the host side, using devices like Tyan’s EXCELLENT Transport GT28 (B2935) with AMD’s quad core Opteron processors allow for rack dense ESX clusters to be created that can scale to (get ready for it): 160 physical sockets/640 cores per 40U rack and 320 Gigabit Ethernet ports. I also forgot to mention that within these 1Us, you can install low profile 2port Qlogic QLE2462 4GB/s fibre cards to allow for multi-protocol attached storage to be used. *hint, hint* I think this would be a GREAT platform for the next EMC Celerra.
c.) Use different storage media. By “different storage media,” I am referring to the availability of SLC/MLC flash drives and the pervasive use of 2.5″ fibre/SAS drives within the data center. I’ve already waxed eloquent before on the merits of using 2.5″ drives (lower power consumption, less moving parts, typically faster access times than comparable 3.5″ drives, etc.) and I’m anxiously waiting to see if EMC will adopt these drives for their arrays. With 2.5″ drives coming close in platter densities (500gb 2.5″ SATA drives are already available in the market), I think there is less of a reason to continue to use 3.5″ drives for nearline storage. Flash, on the other hand, while available in smaller quantities, takes the speed and power equation to a whole different level. I’ll let the Storage Anarchist explain the details:

“As you’ve probably read by now, the STEC ZeusIOPS drives themselves are in fact optimized for random AND sequential I/O patterns, unlike the lower cost flash drives aimed at the laptop market. They use a generously sized SDRAM cache to improve sequential read performance and to delay and coalesce writes. They implement a massively parallel internal infrastructure that simultaneously reads (or writes) a small
amount of data from a large number of Flash chips concurrently to overcome the inherent Flash latencies. Every write is remapped to a different bank of Flash as part of the wear leveling, and they employ a few other tricks that I’ve been told I can’t disclose to maximize write performance. They employ multi-bit EDC (Error Detection) and ECC (Error Correction) and bad-block remapping into reserved capacity of the drives. And yes, they have sufficient internal backup power to destage pending writes (and the mapping tables) to persistent storage
in the event of a total power failure.”

In any case, these are some quick notes from me this AM. Definitely am looking forward to delving into the Tyan GT28/AMD Quad Core stuff in the next few days.

Happy friday!

Technorati Tags: green, datacenter, AMD, EMC, storage, virtualization, SAS, SATA, flash, SLC, MLC, DiskXtender, Centera, VMWare, ESX, Opteron, Qlogic, Celerra, Tyan, STEC, SSD

On Iomega (and other musings)

So, for due diligence purposes, I’m going to remind you to read that little disclaimer stuck in the upper right hand corner of this blog. Since that little bit is over with, let’s get on with the rest of this blog.

DailyTech – EMC Walks Back to Iomega With Revised Offer for Acquisition

If you read the above link (and about 15,000 other links that you can find over @ Google News on the same subject), you’ll see that EMC is starting the “takeover dance” with a little company called Iomega. Iomega, if you remember, was that fiesty little company that tried to get rid of the floppy by introducing the Zip drive, etc. Then came the Jaz drive, the Rev drive, and suddenly, Iomega was basically buried by the pervasive optical drive market. People didn’t care about “hard drive like speeds” on the Jaz and Rev. They were awkward, required specialized hardware (vs. CDs/DVDs which were pretty much record/play anywhere) and they were priced beyond parity with CD/DVD burners.

Flashing forward, they decided to dilute their strengths by playing with NAS and other DAS storage technologies, again at disadvantageous pricing levels and weaker performance than, say, whitebox vendors like Enhance Technology, D-Link, Netgear, Linksys, etc. All in all, they quickly relegated themselves to market followers from a leadership position.

So, why is EMC going after Iomega? Surely they’ve studied brand recognition and penetration and have realized that beyond pitiful market performance (sales and revenue) that there is a certain air of distinction when “Iomega” is heard. For similar comparison, I guess, you could look at the purchase of Linksys by Cisco and how that has managed to maintain profitibility without dilution of the core branding. Conversely, you could look at AMD’s purchase of ATI Technologies and the struggle they’ve had to maintain profitability and market share since then. However, I’m going to come right out and say it: I don’t think that this move is at all a good idea. Here are a couple of points where I think this all falls apart.

a.) EMC prides itself on having a solid portfolio of performance oriented hardware. Some have come through acquisition (Clariion) and some are “homegrown” (Symmetrix). To that end, Iomega’s current position of whiteboxing other people’s hardware seems to be almost a tertiary acquisition of IP and too far down the line to really add value to the overall EMC product portfolio. Almost contradictory to the overall mission of EMC to manage a customer’s information with both quality and performance. A continuation of a whitebox agreement with Intel (a name almost synonymous with “performance” and “hardware”) would do much better.

b.) EMC is trying to move into the SMB space using their reputation for enterprise excellence. Why would you purchase an inferior product when you’re trying to strengthen your overall image within the SMB space? EMC Fortress was a good move with EMC solidifying a tailor-made product that the SMB world could affordably attain. Why EMC would want to re-engineer and re-invent the wheel with Iomega is beyond me. It’ll do more harm than good. They’ll need to redo their software (which is terrible), redo their hardware (which is terrible), and basically restructure the company into an EMC Jr. (time, expense, and…well, profitability dips doing that.) Again, for comparison purposes, Cisco, when it took over Linksys, gained a company with a solid reputation, a solid and profitable portfolio of products, and had little work to do to re-brand and promote.

c.) EMC has better purchasing options available in the SMB focused storage field. Looking at Iomega as a strictly “dilution” oriented brand purchase, what else is out there for purchase? I can think of a few different companies and alliances:
i.) Intel. Not a purchase but an alliance. Take Fortress one step further. Get into the hardware with them. Intel makes the core logic cpus and ASICs that we use (along with many others) and knows how to maintain profitability in a variety of key verticals.
ii.) Buffalo Technology. Buffalo has a VERY good reputation amongst SMB users that I’ve dealt with and again, has a VERY diversified portfolio of storage and IP based products. It’s a PERFECT fit within the EMC market. It also has international reach.
iii.) Enhance Technology. Enhance is an awkward bird when it operates by itself. However, if it had the correct strategy and corporate guidance, it would have tremendous reach. Not only are its product EXCEPTIONALLY well made, they have performance and OEM designs taken care of already. With EMC’s corporate disk alliances (Fujitsu, Hitachi, Seagate), it would be incredibly easy for EMC to incorporate and re-brand their products into EMC SMB.

Anyhow, these are my thoughts for now. Again, reference the disclaimer above as these are MY opinions, not EMCs.

cheers,

Dave

Technorati Tags: EMC, EMC Fortress, Fortress, Iomega, Symmetrix, Clariion, Enhance Technology, Buffalo Technology, Intel, Storage