This is probably a good idea. A company needs a single name, and NetApp is what lots of people (even me) have long called the company. Plus, it’s never good to have your company name be the same as one of your products, at least when you make more than one. And NetApp has lots of different products, many of which are not network appliances…

They’ve added a new logo, too, which ironically looks like a thick blue dolmen to me, but was probably supposed to evoke a door and the letter, N. I always liked the old round peg in a round hole idea, myself… But then again, I always kinda liked yellow and purple and silver storage devices, too!

Remember the old days, when it was Apple Computer, HP still stood for Hewlett-Packard, Sun for Stanford University Network, and EMC for Evil Machine Company? (Just kidding, guys, I know it was Egan, Marino and Einstein’s equation…) But the world will end if IBM ever changes its logo!

Update: More coverage:

• Marc Farley liked the old logo better
• Robin Harris is wined and dined by NetApp in New York – where’s my invite, guys?
• Rajeev Karamchedu thinks it looks like an IBM lego house…
• Zerowait calls it a staple and compares it to New Coke – Ouch!
• The Register is even less complimentary…

Image by Andrew Dunn courtesy of Wikimedia Commons, cc-by-sa-2.0

© sfoskett for Stephen Foskett, Pack Rat, 2008. | The Artist Formerly Known As Network Appliance
This post was categorized as Computer History, Enterprise storage. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

In the interests of all, I’d like to point out two delicious sources of VMware storage protocol wisdom:

An internal paper from VMware’s performance folks titled Comparison of Storage Protocol Performance shows that, as expected, Fibre Channel has the lowest CPU overhead and best overall throughput. But, no surprise to anyone who’s tested alternatives, iSCSI and NFS also work pretty darn well! And you can knock that extra CPU load right down to the FC level with an iSCSI HBA.

Next up is a best practices paper from Network Appliance that is chock full of VMware storage goodness. If you’re curious about the potentials of NFS storage for VMDKs, this paper is a must-read! I’m pretty impressed with what VMware over NFS has to offer.

By the way, my next seminars are June 24 and 26 in Atlanta and San Francisco, respectively. I’ll also be presenting some related content at Storage Decisions in Chicago in May and Toronto in June.

Update:  Marc Farley talks back – isolate your networks, people!

© sfoskett for Stephen Foskett, Pack Rat, 2008. | Which Storage Protocol For VMware?
This post was categorized as Enterprise storage, Virtual Storage. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

Overall, I was impressed with many of these products this last year, but certainly CommVault, HDS, ProStor, Xsigo, and Akorri (winners all) stood out with important products. I too was surprised to see the absence of some heavyweights: Symantec’s strength in archiving and backup didn’t net them an award, and both Emulex and QLogic were skipped for the 8 Gb FC market. EMC was overlooked, too, but I’m certain that will be remedied for 2008 after the flurry of excellent new products announced these last few months.

Backup Software

• Winners: CommVault’s Simpana 7.0 – an impressive product that brings a very wide range of data protection features in an integrated package. They’d be in my top three! Next was the intriguing RecoverGuard from Continuity Software, a product which intrigues me. Finally, Yosemite’s amazingly priced FileKeeper Professional took bronze – I definitely would have put this up there, too!
• Strong contenders: EVault’s solution would have been my pick since it effectively tackles a real need – mobile and SMB data protection! Also, Symantec’s strong NetBackup 6.5 upgrade deserved more attention.

Backup Hardware

• Winners: NEC’s HydraStor HS8 isn’t a platform I’m familiar with – I’ll have to take the judges at their word! But I can’t argue with silver and bronze winners, Copan and ProStor’s cool RDX – both are truly revolutionary products and deserve the spotlight.
• Strong contenders: I would definitely have added Gresham’s Clareti VTL to the list – it’s much more than “just a VTL” and I hope it gets some more attention!

Disk and Disk Subsystems

• Winners: Seagate’s Momentus 5400 FDE.2 deserves considerable attention, bringing built-in hardware encryption to the mobile data market – every mobile disk should have this technology! NetApp’s FAS2000 (silver) is nice enough, but I was much more impressed with the bronze-winning USP V from HDS this year.
• Strong contenders: I would have given the USP V the top prize for 2007, but I can’t argue with the little Seagate disk. The AMCC 3ware 9600 RAID adapter deserves some attention, too.

Storage Management Software

• Winners: Akorri’s BalancePoint (gold) impresses me much like bronze-winning Onaro’s Application Insight does. Both make the big leap from “storage” to “application data”, and both are worthy winners. I’ve never used the silver-winning Finisar NetWisdom product, but it looks like others are impressed with it.
• Strong contenders: This was a crowded field, and Symantec, WysDM, Quantum StorNext, and the rest probably came close. I agree with the Akorri placing, but would have picked one of these instead of the potentially mis-categorized Finisar product.

Networking Equipment

• Winners: Xsigo took the top honors with their InfiniBand-based VP780 platform. This is truly a next-generation product, and it is getting serious attention and traction, and deserved a spot on the list! Riverbed’s excellent Optimization System (silver) also deserved its ranking, but I’m not familiar enough with the Storwize product to know if it’s truly bronze-worthy.
• Strong contenders: I would have tipped either QLogic or Emulex‘s 8 Gb offerings for a spot. Despite my jokes, 8 Gb FC is an important element of the modern SAN and both companies have carved out a compelling product, but apparently neither shipped in volume until this month…

© sfoskett for Stephen Foskett, Pack Rat, 2008. | Reacting to TechTarget’s Storage Products of the Year
This post was categorized as Enterprise storage. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

© sfoskett for Stephen Foskett, Pack Rat, 2007. | Yowza! VMware is a Rocket!
This post was categorized as Enterprise storage, Virtual Storage. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

Tom’s Hardware, one of my favorite nests of geeks, tested a pair of these 32 GB SanDisk drives and came out with some interesting findings. Notable among their findings is the fact that the pathetic random write throughput makes SSDs totally unsuitable to server (read enterprise storage) applications. How does 40 I/O operations per second strike you? Yes, that’s 40, and it’s due to the inherent nature of NAND flash memory and its organization in SSDs. RAID helps some and hurts some – it spreads the load, improving performance, but makes sequential writes even less likely.

But these drive mechanisms aren’t meant for this kind of load – they’re meant for laptops and gamer PCs which heavily lean toward random reads, an area in which solid state disks excel. “Boot Windows in half the time!” (unless you’re comparing the SSD to an actual high-performance disk)

There’s also a little issue of reliability. Well, actually, let’s say longevity. You see, a NAND flash cell is still good for only about 10,000 writes before it becomes a lump. This isn’t much of a problem in a thumb drive or iPod since these don’t actually get written to all that much. Plus, all current SSDs include “wear leveling”, a strategy that maximizes write longevity by moving blocks (on write) to less-used flash cells. So a bigger flash device actually lasts much longer than a smaller one under the same I/O profile because all those extra cells can give their little lives to save your data.

This longevity issue isn’t just academic. Ask anyone with an Unslung NSLU2 booting from a flash drive (yeah, including me) and you’ll hear about failed thumb drives after a year or two of use. No big deal when you’re talking about a $15 item, but what about enterprise storage? I guess EMC and IBM wouldn’t mind forcing you to replace your enterprise storage media every year or two, but how will you feel about it? And what if the NAND was soldered to your motherboard when it failed? Makes Apple’s sealed iPod/iPhone batteries seem trivial, doesn’t it?

Then there’s the issue of cost. A 32 GB SATA SSD drive runs $400 retail, which is about how much you’d pay for a 300 GB SATA 2.5″ laptop drive (if you didn’t get a good deal). In other words, it costs about 10 times more than a comparable spinning drive on a per-GB basis. Lots of companies are investing in flash (iPod effect, anyone?) but NAND prices are not promising to overtake disks any time soon. So we’re left with an exceptionally flawed product.

Of course these drives have no cache – it’s irrelevant, say the manufacturers, in a solid-state device. But write cache might actually improve random write performance substantially, especially if it was backed by a super-smart algorithm to maximize sequential I/O in the same way that Network Appliance’s WAFL optimizes RAID-4.

And if this software also included the wear-leveling smarts, things would be even better. Imagine optimizing writes to kill a single NAND module quickly, sparing the rest of the array. Think tiered storage for longevity instead of cost – frequent write I/O operations go to the sacrificial cells and longer-lasting ones are destaged to the “permanent” ones. Makes media replacement much more palatable, doesn’t it? Add in some smarts and a sizable write-back cache to keep the really transient writes off the flash entirely and you might have something there.

As for cost, consider Mark Lewis’ recent posting about OLTP versus “web” data. He’s telegraphing EMC’s playbook for SSD – smart tiered storage that places small amounts of OLTP data on smart NAND and everything else on regular disks. Sounds workable to me!

One more point to make. Rumor has it, disk giant Seagate is thinking of snapping up memory specialist, Micron. If this isn’t a sign that solid state tech is becoming important to the storage component industry, I don’t know what it is!

So I’m still not too positive on SSD technology. It’ll always be more costly, and the current offerings are woefully inadequate. But I can see a way to make it work in enterprise storage, and I see signs that the big companies are trying to do just that. Wake me when the train arrives, ok?

© sfoskett for Stephen Foskett, Pack Rat, 2007. | Commercial SSDs Are Here?
This post was categorized as Enterprise storage, Terabyte home. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

If it’s true that “stuff”, like gas, expands to fill all available space then at least I now have more space for my stuff. I’ve been burning up the Lowe’s parking lot trying to fill it though…

ObStorage: This reminds me of a trick a NetApp administrator friend used to pull on his users. He would adjust the size of the snap reserve to make his filers always look almost full. He swore that this kind of “backpressure” was the best way to control injudicious use of storage space by end users. Whenever he added a new disk shelf, he would immediately allocate the whole thing to snap reserve, and would then only open up 100 MB or so at a time! This smoothed out his provisioning, no doubt. Kind of anti-thin provisioning!

© sfoskett for Stephen Foskett, Pack Rat, 2007. | Back from the All-Star break
This post was categorized as Enterprise storage, Everything, Personal. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

I’d like to kick off a series of articles with an exploration of a key piece of storage technology, the open systems NAS array. Now, lots of people think that NAS is a new development, but this is not so. In my research, I’ve come to the conclusion that NAS predates SAN by a few years at least, and its history is linked to the development of open systems servers, too!

Let’s start with some basics. I’m assuming that NAS is defined as the sharing of files (rather than blocks) over a high-level protocol. NAS generally addresses offsets within files within folders, and we usually encounter it today in the form of CIFS or NFS servers, which operate over the familiar IP protocol and Ethernet networks.

This was not always the case, of course. The earliest file servers I could find were created at Stanford using Xerox Alto servers, and headless file servers were named and in place by 1979, according to Byte magazine. Certainly, development of the concept of a “server” and file server in particular was helped by the introduction of XNS around 1981, as it included RPC functionality.

Novell took this concept and ran with it, transforming XNS SPP into IPX/SPX and introducing NetWare in 1983. It’s safe to say that NetWare was the first file server software, at least in the open systems world.

But there was another heavy hitter in town – 3Com. These days, it’s easy to forget just how important this company was back then, but the networking and storage world would look very different without 3Com! It was founded to exploit Xerox PARC’s Ethernet protocol, and like Intel today spent much of its first decade pushing networked applications into the market.

3Com developed a network server operating system of their own on top of DOS – 3+Share. Over two decades, this product would evolve into LAN Manager, SMB, and CIFS!

But 3Com released a hardware product, too, and this is critical to our exploration of the storage industry. The 3Server was based on the Intel x86 architecture and booted MS-DOS, but was not a PC. It had no provision for a “head” (keyboard and monitor), and was managed remotely over the network. It included seven disk drive slots from its 1985 introduction and included software to manage these disks and present storage over the network. Let’s see – headless dedicated server with disk slots running a proprietary file serving OS. Sound like a storage array to you? Me too!

Although it originally supported XNS over Ethernet and AppleTalk, Token Ring support was added quickly. The 3Server (like NetWare) also supported network applications, but it was its storage protocol that had the most impact. 3Com worked with IBM to develop a successor to 3+Share, which IBM called LAN Manager and 3Com called 3+Open. This was based on OS/2 and was handed over to Microsoft in early 1991 as 3Com refocused on network infrastructure.

So who knows of an earlier storage array in the open systems world? I’ll cover Auspex/NetApp, EMC, and the rest in future installments of Storage History.

© sfoskett for Stephen Foskett, Pack Rat, 2007. | Storage History: The 3Server
This post was categorized as Computer History, Enterprise storage. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

Consider that glorious American culinary innovation, the buffet. People love buffets. But a friend in “the biz” tells me that restaurants have a love/hate relationship with them. On the one hand, people tend not to eat any more at a buffet than they would if just ordering an all-inclusive meal (really!), though a la carte menus do tend to cut into food consumption. Plus, restaurants can charge much more per head at a buffet than any other menu system, so there’s guaranteed revenue per customer. But people tend to waste more food at China Taste Buffet than at Chef Lo’s, and they tend to gravitate to the expensive spare ribs and general gao’s instead of the cheap lo mein.

But the important thing is that customers love the buffet experience. They feel like they’re getting a bargain when they gorge on “free” shrimp, even if they aren’t really getting any more for less. It’s in our nature to want something for nothing.

So what about storage? Which model is best for customers – EqualLogic’s philosophy of bundling all features with hardware or NetApp’s old stance that customers should pay for each feature they want? Clearly, bundling helps when a customer plans to use a lot of features, since replication and application integration can easily double the cost of a mid-range storage array.

But there’s the confounding factor that most people don’t use many storage features. Back in the ’90s it was “common knowledge” that more than half of all EMC TimeFinder licenses were “shelfware”. Of course this had something to do with how much of a pain it was to integrate with applications, but the sales people were happy to bundle it with every Symmetrix purchase! The question is whether people on the whole benefit more from bundling or un-bundling…
Perhaps more people would use advanced features if they were free. I guess this is the goal of the storage upstarts – get folks to love the integration provided by their bundled features so they won’t want to switch to another array as their needs grow. I for one would love to see more people integrate snapshots and replication into their storage environments to improve recoverability, so I’m all for it.

© sfoskett for Stephen Foskett, Pack Rat, 2007. | NetApp heads to the buffet
This post was categorized as Enterprise storage. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.