EMC VFCache is a Simple Read Cache

I was not pre-briefed on this product, and I’m not all that thrilled at the prospect of attending a launch webinar, so what you read here is based on my own research and reading of the available information as of this morning.

When EMC announced Project Lightning last year, company insiders expressed surprise to me. It seems that many had never heard of the project, and those that had didn’t think it was far enough along to be announced. I didn’t even bother to write about the Project Lightning announcement at the time. But today EMC unveils the production product that came out of Project Lightning.

VFCache is a filter driver that caches writes

EMC VFCache appears to be a simple and straightforward offering:

1. A PCIe SSD from Micron or LSI sits in the server and acts as a read cache to accelerate performance
2. EMC software also runs on the server, snooping on I/O and filling the cache with relevant data

There’s not a lot more to the product than that. EMC will sell the PCIe SSD and bundled software as VFCache, and will no doubt market the heck out of this product. Perhaps the only novel twist is the so-called “split-card” mode, which allows the card to act as a write cache. But EMC only supports this for transient “throwaway” data with direct attached storage (DAS) as a backend. There’s no way a conservative, enterprise focused company like EMC would risk sanctioning a writeback cache with no redundancy or data protection features.

VFCache uses a filter driver installed in the VM guest

Perhaps the biggest limitation of the initial VFCache offering is its limited applicability to enterprise server virtualization environments. VFCache uses a filter driver installed in each VM guest, and includes no hypervisor drivers though there is a vCenter plug-in. This makes VMware vMotion very tricky, involving scripting to remove and re-add storage. This means VMware SRM will not easily work, and there is no support for clustering, either.

This is no surprise, since VFCache appears to the host as a local storage volume (AKA, a disk drive or LUN) which would disappear if a virtual machine is moved to another server. Virsto solved this problem by virtualizing storage presentation to the hypervisor, and Fusion-io’s ioTurbine software does not interfere with vMotion either. EMC will likely go in this direction in the future, but it’s a big hole in the product for now.

You might also like reading Micron Bursts Into the PCIe SSD Market to learn more about the card EMC is using

The News: EMC Is in the Host-Based Storage Business

The primary use case for this product is server I/O acceleration. This is desperately needed, as applications and servers are rapidly outrunning the capabilities of conventional storage arrays. EMC and other legacy array manufacturers initially tried to address this I/O imbalance with tiered storage and in array caching. Indeed, these technologies are fairly effective at accelerating the performance of conventional disk storage arrays.

But flash manufacturers like Fusion-io (not to mention Micron and LSI) absolutely demolished storage array performance with their in-server offerings. EMC faced the prospect of losing out on the high-performance storage market. EMC simply could not allow their bread-and-butter enterprise customers to look elsewhere for strategic, high-performance storage for high-profile applications.

VFCache gives EMC salespeople a silver bullet when customers demand maximum performance, but this launch may not spell doom for the flash startups. For one thing, it legitimizes host-based flash cards as a viable component of enterprise storage architectures. It also opens the door to comparison between SAN storage and non–SAN alternatives that go well beyond what EMC is currently offering.

Shared Flash Storage Is on Deck: Project Thunder

"Project Thunder" will externalize the PCIe flash cards over a high-performance "Server Area Network"

As part of the VFCache introduction, EMC is also talking about Project Thunder, a shared version of VFCache. At the very least, thunder will allow multiple servers to access a shared pool of flash cache. This should allow VMware vMotion and DRS to function, and could be much more than that.

EMC could build a high-availability, high-performance all-flash storage array that may even use InfiniBand as an interconnect. The new Nimbus Data E-Class storage array matches this description perfectly, and their CEO tells me that performance over InfiniBand is indeed comparable to in-server PCIe flash cards. It seems logical for EMC to enter this market, if only to disrupt the momentum of Fusion-io and the rest of the all-flash storage upstarts.

Read more about the Nimbus E-Class: The First Big, Redundant, All-Flash Enterprise Array

The only fly in the ointment here is the recent consolidation of the InfiniBand market. Mellanox bought Voltaire, and QLogic sold out to Intel, putting that protocol on tenuous grounds. Perhaps 40 or 100 Gb Ethernet will emerge as a viable alternative for high-performance connectivity, or perhaps these products will retrench on shared PCI Express instead. Micron recently purchased Virtensys for just such a product, and Xsigo has been making big waves in the area of converged I/O as well. The market clearly need something better than Fibre Channel for maximum performance storage, even if InfiniBand isn’t it.

Stephen’s Stance

EMC VFCache (née Project Lightning) is a fairly simple offering: A server-based PCIe flash card that acts as a read cache with no integration with storage arrays or hypervisors. But EMC’s entrance into the host-based flash storage market is a powerful demonstration of the wave of disruption caused by flash-based storage and high-performance computing. Although I am not all that impressed with the product itself, I would be distressed if EMC had not introduced it.

VFCache illustrations are copyright EMC Corporation and are used here with permission

More solid, independent VFCache coverage:

• EMC Enters The Market With “Me Too” Flash Products (Chris Evans)
• Cache Splash and Complex is the new Simple (Martin Glassborow)

© sfoskett for Stephen Foskett, Pack Rat, 2012. | EMC VFCache (aka “Project Lightning”) Is One Small Step, But an Important One
This post was categorized as Enterprise storage, Gestalt IT, Virtual Storage. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

How could a company as mighty and diverse as HP have had so many issues with executive management?

HP stumbled mightily in 2011, and it had nothing to do with product or people. Even sales remained strong, though the PC business is changing. HP’s mighty stumble was a crisis of confidence due to a chain of shenanigans at the very top. This culminated with the short reign of Léo Apotheker, leaving HP to reassure the market of its strategy.

HP And the Enterprise IT Industry

Sometimes, it’s hard to get a sense of scale when talking about very large things. How big is the solar system? How far away is the nearest star? The same is true of earthly things, exemplified by popular misconceptions about the global financial crisis. It’s difficult for people to understand just how much money a trillion dollars is.

In my little world of enterprise storage, it’s difficult to reconcile “big storage” players like EMC and NetApp with “big storage and everything else” players like HP, Dell, Oracle and IBM. Sure, EMC and NetApp lead the pack in terms of market share, but they’re nowhere near as large as the integrated players. HP has more than 7 times the revenue of EMC, which makes 3 times more than NetApp.

This old slide might need updating, but you get the picture...

HP is an incredibly diverse company, dominant in the PC, printing, and blade server market and top 5 just about everywhere else, including networking, services, and enterprise storage. And HP has nearly 325,000 employees, all working to move the company forward in one direction or another.

NetApp is a motorcycle, with one drive wheel pushing it forward at high speed; HP is more like NASA’s shuttle crawler-transporter, a 16 motor mammoth. Single-purpose companies can be agile, but they can also be derailed by market downturn or technological shift. Storage specialists like NetApp continually try to innovate and acquire to keep themselves vital, while larger companies like Cisco and EMC try to diversify while maintaining market leadership. HP doesn’t need to try; it is diverse.

HP Is a Very Large Thing

I watched HP's rise in the server market of the 1990's

HP has long been synonymous with innovation, high-technology, and silicon Valley. I have been an HP customer as long as I have been in IT, and watched as they integrated technology from Apollo and Convex in the 1990’ s. The server products that resulted became the dominant UNIX platform, but HP’ s innovative storage concepts didn’ t take the market by storm.

After HP merged with Compaq (which brought Tandem and Digital Equipment Corporation), the company vaulted ahead in the Wintel market and also gained valuable storage expertise. Throughout the last decade, HP was firing on all cylinders and dominant in nearly every arena. The company owns half the blade server market, is tied for first in servers, leads in PCs and printers, and is a contender in networking and storage. It’s simply impossible to say what HP is in a single sentence.

HP storage has an extremely broad product range, which management is working to reconcile. Acquisitions of LeftHand, Ibrix, and 3PAR gave HP storage shot in the arm to be sure. An injection of startup mojo has energized the marketing and product groups within HP just when the company needed it. HP’ s market share has grown somewhat as a result, though not as much as the hyper-focused NetApp. HP networking similarly took on 3Com, bedeviling Cisco in the Ethernet switch market.

The Executive Soap Opera

It takes a lot to bring a mammoth to its knees, but a shot between the eyes usually does the trick. While the many engines of HP push it forward, the company’s upper management has seemed, at times, suicidal. Business schools could design an entire curriculum around the folly of Patricia “I spy” Dunn and Mark “penny-pinching” Hurd. Who would think that HP management could top this?

From August 2010 through September 2011, HP dominated IT headlines in completely the wrong way. The board wanted a change, and selected Léo Apotheker to transform HP. But it was a soap opera from the very start, with Oracle hiring Mark Hurd and sending Apotheker into hiding among accusations of corporate espionage while at SAP.

The new CEO didn’t seem to understand HP at all, though he was intent on steering it in a new direction. Apotheker set about dismantling HP’s consumer and end-user businesses, killing Palm/WebOS and threatening to sell off the PC business. The company was to focus instead on enterprise computing, but these drastic moves spooked the entire industry.

It wasn’t long before HP’s board struck again, with a shake up at the hands of Ray Lane and former eBay CEO Meg Whitman brought in to replace Apotheker. The first order of business for the new HP executive team appears to be reassuring the entire world that management has not gone completely insane.

Stephen’s Stance

To an outsider like me, the most disappointing thing about HP’s mighty stumble is that it has nothing to do with the people who really make the company what it is. I have met many creative, hard-working individuals in HP’s storage, server, networking, and printer groups, and they could not be more different from the executive soap opera. I only hope that this new board and CEO will bring some stability and let HP cruise forward once again.

Disclaimer: HP has sponsored the Tech Field Day events which I organize, and has on occasion invited me to attend events at their expense.  But I do similar work with nearly every company in the IT industry, and this piece is my own opinion.

© sfoskett for Stephen Foskett, Pack Rat, 2012. | HP’s Mighty Stumble
This post was categorized as Computer History, Enterprise storage, Gestalt IT. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

VASA allows a "provider" application to tag vSphere storage with a "capabilities" string

The list of VMware storage integration features in vSphere is growing at a rapid pace. From backup to enterprise array offload, VMware is staking their claim. But information about one new storage feature in vSphere 5 has been scarce: The true nature of the Storage API for Storage Awareness (VASA) is only just beginning to be revealed.

VASA: Born of Necessity

vSphere has some amazing automated storage mobility features. The ease of moving data between LUNs and even arrays with Storage vMotion is a revelation to those of us unlucky enough to have used manual migration methods in the past. And VMware has automated this with the new Distributed Resource Scheduler (DRS) capability, allowing vSphere to make its own decisions about data placement.

But this kind of dynamic movement can hurt as well as help. What if an administrator moves a VMDK from high-performance to low-performance storage? Server responsiveness would suffer, perhaps even resulting in an application outage. And how could Storage DRS avoid doing the same thing?

VMware realized they needed some mechanism that could “tag” a storage resource with its capabilities. This is the real reason for their creation of Profile-Driven Storage and VASA.

VASA Does One Thing

In vSphere 5, VASA is incredible simple: It is a basic protocol for vSphere to request a “capabilities” string regarding a LUN or NFS share from a “provider”. The content of this response, and indeed the form that this provider will take, is left up to the individual storage vendors.

VASA Providers can take many forms. Dell is reportedly developing a vCenter plugin to report capabilities. EMC and NetApp will use a software application that runs on a virtual or physical server. In all cases, the software uses a proprietary call to the storage array, in effect relaying and translating the VMware request.

The VASA “Capabilities” string is undefined as well. Most vendors use a list of technical attributes as their response string, and each has adopted their own strategy about how to present information:

• Dell’s EqualLogic returns a comma-separated list like “RAID, SSD, REPLICATED”
• NetApp’s provider uses semicolons, as in “Dedupe; Replication”
• HP’s is very detailed, with a schema specifying Drive Type, RAID Type, Provisioning Type, VV Type, and Remotecopy
• EMC appears to have a standardized set of much-shorter tags, like “Performance”, “Multi-Tier”, “Capacity” and such

These are passed through to vCenter, where an administrator can decide how to interpret them.

VASA Today and Tomorrow

In its simplest form, VASA capabilities tags enable both manual and automatic storage placement features to respect tiered storage policies. This is important to maintain proper system performance and availability. In the future, I expect more advanced VASA providers, perhaps even integration of the providers into vCenter plugins. I also look forward to a standard capabilities tagging schema and smarter handling of returned tags. Right now, for example, EMC’s Clariion and Symmetrix lines both tag LUNs as “Performance”, but these are obviously not equivalent.

© sfoskett for Stephen Foskett, Pack Rat, 2011. | What is VMware VASA? Not Much (Yet)
This post was categorized as Enterprise storage, Features, Virtual Storage. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

Like Oedipus, VMware has a tragic destiny

Later this week, I’ll be presenting a session at Storage Expo NL which I imagine will be quite controversial: “VMware as Oedipus: How Server Virtualization will Change Storage Forever“. That’s right, I’m suggesting that VMware is the biggest threat to traditional enterprise storage, and that the company might just bump off its own parent (EMC) in the process. What the heck am I thinking? Read on!

Let’s Talk About Oedipus’ Fate

Most people have only a passing knowledge of the tragedy of Oedipus: he killed his father and slept with his mother, or something like that. But the classical plays about Oedipus revolve on a theme with much broader relevance: How do free will and fate interact?

Both Oedipus and his father, Laius, set up their shared tragedy in response to related prophecies. In trying to avoid their fate, both bring it about. This is a common theme in Greek tragedy and in life more generally.

Before Oedipus is born, his father, King Laius of Thebes, is told by an Oracle that he is doomed to perish by the hand of his own son. In response, he sends the infant away to be killed. Unable to do the deed, Oedipus’ mother, Jocasta, send him away with a servant who leaves him on a mountaintop. A shepherd rescues him and takes him to the childless King of Corinth, where he is raised.

On hearing rumors that he is not the biological son of the king, Oedipus goes to the Oracle at Delphi for advice. But the Oracle tells him that he is destined to murder his father and mate with his own mother, a repulsive fate indeed. To avoid this, Oedipus leaves Corinth and heads to Thebes.

On the road to Thebes, Oedipus meets an unknown man and the to quarrel over who has the right-of-way. Oedipus kills the man and continues on, solving the riddle of the Sphinx and freeing the kingdom from her curse. His reward is kingship and marriage to the Queen. The prophecy is fulfilled: The man on the road is King Laius, and Oedipus’ new wife is his own mother, Queen Jocasta.

The plays of Sophocles, Aeschylus, and Euripides explore the implications of the story and the impact that the revelation has on the characters involved. The most important question raised is that of fate versus free will: Could Laius have made his own fate worse by sending away his infant son? And could Oedipus have avoided the prophecy by choosing thought over action? Sophocles decided that fate could not be avoided, no matter what the motivations of the individuals involved.

The Fate of VMware In Enterprise IT

This is the prophecy that I bring to the audience at Storage Expo NL: As VMware adds both internal capabilities and external integration with storage devices, they are destined to come into conflict with the storage industry in general and their corporate parent, EMC, in particular.

Storage is one of the most critical gating factors to the success of server virtualization, so it comes as no surprise that VMware is rapidly innovating in this area. Integrating and developing snapshot, replication, thin provisioning, and other features in VMFS enables everyone to have advanced storage functionality, regardless of which storage device they use. In this way, VMware is already causing many users to forego an enterprise storage array purchase.

But VMware is owned by the King of Storage, EMC, and surely recognizes the benefits that enterprise storage arrays can bring to the hypervisor. Therefore, the VMware engineers are simultaneously attempting to add tighter integration to SAN and NAS storage systems. Truly, VMware does not want to destroy the enterprise storage industry anymore than Oedipus wanted to marry his own mother!

Now comes the tragedy. As VMware innovates on storage, they run up against the inevitable conclusion that current protocols and arrays are poorly suited to the sort of “multiplexed” random I/O that characterizes virtual server workloads. They’ve already noticed this, and responded with vCenter plugins, VAAI, VASA, and a proposal to add demultiplexers to enterprise storage arrays. Storage array vendors are scrambling to keep up with the rapid pace of innovation triggered by VMware’s integration features.

The irony of all this is that this ever tighter integration serves to anonymize and homogenize enterprise storage devices. If all array management is done within vCenter, and only VAAI-supported storage features are used, brand loyalty is no longer required. After all, one storage device is just as good as the next if all offer similar levels of capability and performance. The likely beneficiaries of the nomination of enterprise storage are the vendors of complete solutions: HP, Dell, IBM, and (yes) the EMC/Cisco combination.

But EMC and Cisco are not low cost providers. Their current success is based on the combination of innovation, integration, and support. But the big three server vendors are hot on their tail in all three areas and are better equipped to add value pricing as a sweetener. Again, if an HP server and storage combination works just as well a vBlock, with identical features and management in the VM environment, why not buy it?

Stephen’s Stance

VMware doesn’t want to hurt its parent, EMC, any more than Oedipus desired his own parents’ fate. Indeed, VMware spends an incredible amount of time and effort innovating both internal and external integration features for storage. They do this to meet their own I/O demands, not out of bloodlust or hubris. But like the tragic hero in a Greek play, VMware is destined to anonymize and homogenized enterprise storage, and this will drastically affect the future of EMC and other pure storage vendors.

Some Responses:

• Fate, Free Will, Virtualization, and Storage
• Is VMware forcing the homogenization of enterprise storage?

© sfoskett for Stephen Foskett, Pack Rat, 2011. | VMware as Oedipus: How Server Virtualization will Change Storage Forever
This post was categorized as Enterprise storage, Personal, Virtual Storage. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

whith·er – Adverb/ˈ(h)wiT͟Hər/
1. To what place or state: “whither are we bound?”.

VMworld always generates buzz, but news of a major push to change the basic access method for enterprise storage took many by surprise. Extending the work already done with VAAI and VASA, this new development takes VMware storage integration to a whole new level. But the one element of announcement caused alarm for many: VMware’s admission that they would be working with just five major enterprise storage companies to develop this technology. Missing along with the many exciting storage startups is Hitachi Data Systems (HDS), undoubtedly a major player in the industry.

HDS and VMware: Expertise and Partnership

Where is HDS in VMware's roadmap?

Hitachi Data Systems (HDS) is perhaps not as well-known as storage giants EMC and NetApp and server leviathans, IBM, HP, and Dell. But HDS is a major player in the industry with a long history of innovation and expertise in storage and server virtualization.

HDS has lately driven innovation in virtualization of block storage (the VSP), object storage (HCP), midrange performance (AMS), and recently announced they would acquire enterprise NAS contender, BlueArc. Although not quite market leaders, HDS has a huge base of enterprise storage customers and a broad product line from midrange to massive scale.

HDS was right there with EMC and NetApp at VMware’s original announcement of VAAI, even as the mainstream products from IBM and HP lagged months behind. And HDS’ Chief Scientist, Clais Mikkelsen, assured me at their “Geek Day” earlier this year that his company was deeply involved in developing the VAAI specification with VMware. Indeed, VAAI was a major theme of the presentations back in March, with Virtualization Product Manager Michael Heffernan dazzling us with his knowledge of the subject.

You might also like reading Concerning HDS and VMware VAAI Storage Array Support in Plain English

Wherefore Art Thou, HDS?

Now that all that has been said, consider how startling VMware’s omission of HDS was when outlining “VAAI 3.” This is a huge snub for such a major player in the industry with deep expertise and a long history of partnership with VMware. Contrast this to IBM and HP, who were notably absent in many earlier discussions of VAAI, and are still working to bring VAAI to all their platforms. Only HP’s LeftHand and (ironically) Hitachi-sourced XP/P9000 arrays included VAAI plugins from the start. HP’s 3PAR had VAAI too, but HP didn’t have that yet.

Many will likely blame EMC, claiming their influence on VMware (a child company) pushed HDS aside. This same line of reasoning was suggested regarding IBM and HP when VAAI version 1 appeared. But IBM and HP (not to mention NetApp and EMC’s new rival Dell) are at the table this time around, and EMC seems far more concerned by competition from them (not to mention new startups!)

So Where is HDS?

Let me get this straight: WMware is NOT working with HDS on next-gen VAAI storage?

Perhaps this is all some sort of gigantic mistake. Maybe the VMware presenter simply failed to include HDS in his list. Or maybe HDS didn’t choose to get involved this time around, though I can’t fathom why. I put the question to HDS on Twitter over the weekend and hope to hear some sort of answer, though I fear that a convincing response might not come.

The HDS response was unsurprisingly nonspecific (and surprisingly "teen txt-spk")

The only response I got from HDS was a tweet from Claus Mikkelsen stating that HDS and VMware “work all levels to deliver solutions” and that there was “more 2 come.”

What does this mean? I can think of a few possibilities:

1. HDS is way beyond every other storage company, and the “VAAI Cabal” are themselves the odd ones out, trying to keep up with HDS’ “L33T” tech and VMware influence
2. HDS was indeed omitted from the list and possibly the “cabal” and are busy working in the background to make sure they’re included in the future

Personally, the second possibility seems much more plausible.

Update: Some responses to this post:

• Exclusion or Not? (Scott Lowe of EMC)
• VAAI Posturing (Chris Evans)

Stephen’s Stance

If VMware aims to transform storage presentation, and is working with major storage vendors to make it happen, HDS ought to be part of it. Their history, technology, and market position earn them a spot in the “VAAI Cabal” and their omission was a bombshell to industry-watchers like me.

Then there is the other question: What about the startups? Innovation in enterprise storage is often driven by new companies, and VMware would be better served by working with the likes of Tintri, Nutanix, and Fusion-io than the same old major players. But this, as they say, is a topic for a different day.

Disclaimer: I’ve never done business with HDS, but they did fly me to the UK for their 2011 “Geek Day” along with a number of other independent bloggers. I have attended similar events sponsored by HP, IBM, and EMC. VMware, EMC, Dell, HP, and NetApp have sponsored Tech Field Day, and I am currently writing for an online community supported by IBM.

© sfoskett for Stephen Foskett, Pack Rat, 2011. | Alas, VMware, Whither HDS?
This post was categorized as Computer History, Enterprise storage, Everything, Gestalt IT, Virtual Storage. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

Iomega updated the 12-drive rack mount StorCenter device, introducing the px12-350r today

Iomega surprised exactly no one by announcing an updated 12-drive rack mount storage array today. Featuring “Cloud Edition” software introduced earlier this year, the px12-350r also sports mildly updated hardware specs, though still relies on Intel’s “Core2 Duo” CPUs and Gigabit Ethernet. The new device slots in between the desktop px4/px6 line and parent EMC’s new VNXe storage devices.

What’s In A Name?

According to Iomega’s established naming standard, the px12-350r uses the new “Cloud Edition” OS (“px”), features 12 drives, ought to perform better (“350″) than the existing “300″ devices, and is rack-mountable. Comparing it to the previous-generation ix12-300r reveals what Iomega changed and what they didn’t.

What’s new compared with the ix12-300r:

• Iomega stepped up from 5900 rpm “green” drives to faster 7200 rpm disks for more rotational IOPS
• 3 TB drives are now available, for a maximum of 36 TB raw storage, though many customers will likely choose less-expensive 2 TB drives instead
• Solid-state drives (SSD) are now available as an option like the desktop px line, though Iomega doesn’t get automatic sub-LUN tiered storage like EMC’s VNXe line
• The “Cloud Edition” OS includes Iomega’s nifty “personal cloud” replication features, allowing data to be moved to other Iomega px devices
• This new OS also supports SCSI3 persistent reservations, required to support many clustered server operating systems
• Iomega finally rolled out the Avamar deduplicating backup client I spotted last year at EMC World a few months ago, and it will be a featured component of the px12-350r
• On the hardware side, Iomega doubled the system memory to 4 GB, so it should handle more users than the previous device
• The px12-350r still uses Intel’s older Core2 Duo CPU range, though it’s not clear if this is the same 3 GHz E8400 chip
• The ix12-350r still features quad Gigabit Ethernet ports, not 10 Gigabit Ethernet like some competing products from companies like D-Link, HP, Overland, and Winchester

Impressions of the px12-350r

The software is all-new, but the px12 still relies on a Core2 Duo CPU and Gigabit Ethernet

Overall, I’m pleased but not overly amazed by Iomega’s new top-of-the-line device. This is partly due to the fact that I knew what was coming: This product was telegraphed by the px4-300r introduced previously. In fact, the shipping px12-350r doesn’t even offer the enhancements I expected: I suggested it would include Sandy Bridge CPUs and hoped for 10 Gigabit Ethernet.

I still expect a transition to next-generation Core CPUs in the coming year, and boxes like this will likely use that extra horsepower to run additional applications. I’m dreaming of a Sandy Bridge Xeon box with an integrated copy of VMware ESXi and 10 Gb Ethernet – such a device would be awesome for remote offices or small businesses!

Iomega’s “personal cloud” technology is perhaps improperly named for this product. It allows a small or mid-sized business to replicate and protect data to another Iomega px-series device without any monthly fees. The device can also back up to Mozy or Amazon S3, and the integrated Avamar client will be appreciated by larger customers.

This product, like the desktops and the previous-generation ix12-300r, includes PCI Express slots. I had hoped these might be populated with optional 10 Gigabit Ethernet cards (as many competitors offer) but these haven’t turned up yet. Iomega suggests smaller IT groups are still holding off on 10 GbE, perhaps waiting for 10GBASE-T. But 10 GbE pricing is dropping rapidly, and I’m seeing much interest from end users. I strongly suggest Iomega roll out this connectivity option in short order!

It’s interesting that VMware chose to develop their own software for the vSphere Storage Appliance (VSA) in vSphere 5 rather than use this Iomega/EMC LifeLine OS. LifeLine is possible as a VM: It will be available in the lab at VMworld! But this is not available for production.

Stephen’s Stance

Despite a lack of surprises, the px12-350r is still a solid SMB or remote office product. Retail price is $6k-$10k, slotting in below the EMC VNXe3100, and the device’s features reflect this product stratification as well. The consumer-sounding “personal cloud” data replication is actually much more useful in a small-business environment, and I suspect Iomega resellers will be pushing this (as long as they don’t sell some other replication option).

I expect further product updates from Iomega in the coming year, including a switch to Sandy Bridge (second-generation Intel Core) CPUs and the availability of 10 GbE. I’d also like to see the LifeLine software (and Iomega’s recommendations) abandon RAID5 in favor of some wide-striping “post-RAID” for performance and reliability. And integrated automated storage tiering would be great, though this would really show up the “big brother” VNXe line. Hey, a guy can dream right?

© sfoskett for Stephen Foskett, Pack Rat, 2011. | Iomega Introduces the StorCenter px12-350r
This post was categorized as Enterprise storage, Everything, Virtual Storage. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

A Better Way to Dedupe Data: Dedupe Everywhere

You might also want to read When Marketing Becomes Pointless and The Epidemiology of Viral Videos

Stephen’s Stance

If you’ve been following the IT industry and my blog for a while, you probably recall the Mini Cooper stunts EMC pulled earlier in 2011: They packed a bunch of contortionists into a Mini at their New York event, while simultaneously parking a bunch of logo covered cars in front of NetApp headquarters. This was intended to demonstrate something about storage, but the exact point escaped me. In the end, the whole thing seems mean-spirited and pointless.

The Symantec video included here is an entirely different animal. It’s exactly the sort of thing I approve of: A lighthearted look at a serious technical topic with only a gentle poke at the “opposition”. And I bet producing this video was a lot less expensive than hiring an acrobat school! Let’s hear it for social media!

Disclaimer: Symantec is a frequent presenter at Tech Field Day and I have worked with them often on other projects. But this post is my own idea and I’m getting no compensation related to this video. I just liked it and wanted to spread the word!

© sfoskett for Stephen Foskett, Pack Rat, 2011. | How Many Social Media Marketing Campaigns Fit Into a Mini Cooper?
This post was categorized as Enterprise storage, Personal. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

These look to be Teradata Extreme Data Appliance racks

1. In the first photo, we have rows of swoopy racks – at least 30 are pictured. These appear to be Teradata Extreme Data Appliance racks. The gray color and funky doors give them away.


This photo features rack servers and NetApp storage

2. The second photo includes a 1U rack server with six 2.5″ drives featuring “HP purple” drive eject buttons. There are over 100 in this photo, receding to infinity. Looks like HP ProLiant DL360 G7 servers to me.
3. We also have a 2U rack server with 16 2.5″ drive bays. I count over 20 of these. They don’t quite look like HP to me, but could be high-density HP DL380 G7 servers.
4. What looks like a NetApp FAS6200 to me is at the lower right. It looks to be 6U high, compared to the shelves.
5. We also have a number of disk shelves for the NetApp. I’m guessing NetApp DS2246 shelves.


This still from Apple's official video shows more NetApp storage and HP servers

6. In this third shot, we see more HP 1U servers, possibly ProLiant DL360 G7′s as noted above.
7. We have another NetApp filer “head” similar to the one in the second shot.
8. Some sort of 6U beige rackmount device is found in the bottom of two racks
9. We have many more NetApp disk shelves or 1U servers above

Note that these racks contain many 1U Chatsworth “HotLok” filler panels. They’re not very efficiently racked, but perhaps this was due to flexibility or power requirements…

Stephen’s Stance

Whatever Apple bought for this datacenter, they bought a lot of it. Just the hint that they purchased Isilon bumped parent company EMC’s stock a few months back. I imagine Teradata and NetApp might see similar bumps from these images!

It’s also interesting to speculate on the political implications of Apple’s datacenter buys. Would they still buy HP equipment, now that Palm and WebOS are re-emerging as iOS-killer contenders? Would Jobs’ friendship with Larry Ellison indicate a proclivity to buy Oracle or Pillar Data? Maybe the Apple/Cisco Détente would suggest UCS or even EMC? And what about that rumored Isilon buy. Was that displaced by Teradata and NetApp?

Images from MacRumors.com

Note: This has been one of my most-popular posts, and has been referenced a lot by major news organizations. Here’s a list of those who linked and mentioned me. Thanks!

• The web’s watchful eye fixes on Apple’s cloud gear
• Steve Jobs Provides A Look Inside the iDataCenter
• WWDC: Has Apple turned to Oracle for iCloud?
• What Kind Of Gear Is Apple Using In Its Huge New Data Center? (AAPL)
• How Green Is Apple’s iCloud?

© sfoskett for Stephen Foskett, Pack Rat, 2011. | What Datacenter Equipment Is Apple Using?
This post was categorized as Apple, Enterprise storage, Personal. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

VMware's PSA is awash in abbreviations and options

I am often questioned during my Storage for Virtual Environments seminar presentations about VMware’s Pluggable Storage Architecture (PSA). This system is fairly straightforward and concept: VMware provides native multipathing support for a variety of storage arrays, and allows third parties to substitute their own plug-ins at various points in the stack. But the profusion of acronyms and third-party options makes it difficult for end-users to figure out what is going on. In an effort to help, I present here another entry in my “VMware storage features in plain English” series.

Note: I am more of a storage guy than a virtualization expert. I consider myself one of those end-users who have had trouble figuring out what’s going on with PSA specifically, in VMware storage features in general. I welcome comments and suggestions for corrections or improvements to this and all of my articles. Thanks for your help!

Introducing Pluggable Storage Architecture (PSA)

Pluggable storage architecture was one of the major enhancements introduced in vSphere 4. Functionally similar to Microsoft’s MPIO stack for Windows, PSA includes native multipathing support and allows vendors to plug in their own advanced features.

I find the VMware diagram confusing. Is mine more or less accurate and readable?

The ESX kernel (VMkernel) walks down through three layers when communicating with storage:

1. In the top layer, VMware native NMP or third-party MPP software decides which SATP to use, or whether to use the native interface. MASK_PATH also operates at this layer.
2. The SATP layer includes native generic path selection (active/active, active/passive), standard ALUA, as well as allowing third-party plugins (SATP) to override its behavior. The SATP monitors these paths, reports changes, and initiates fail-over on the array as needed.
3. At the PSP layer, software decides which physical channel to use for I/O requests.

There are three types of PSA plugins for vSphere 4:

1. Storage Array Type Plug-In (SATP)
2. Path Selection Plug-in (PSP)
3. A complete third-party multipathing software stack (MPP)

As is the case with VAAI, VMware includes a number of third-party plug-ins in the ESXi install. Users can simply activate many of these according to their needs, though some require additional fees and licensing.

Storage Array Type Plug-in (SATP) List

Storage Array Type Plug-Ins (SATPs) to the VMware Pluggable Storage Architecture multipathing solution for the specific characteristics of the storage array. This is very important, since each storage array design differs substantially in detail and support, especially when it comes to load-balancing and failover between controllers, ports, and paths. So it is critical for VMware to have developed a standard interface to communicate with arrays.

SATPs allow load balancing across multiple paths, intelligent path selection, and over troubled conditions such as “chatter”, when passed rapidly fail back and forth between controllers.

The SATP has critical tasks to perform in the PSA stack:

1. Decide which method of communication to use with the storage (PSA or native)
2. Monitor the health of the physical I/O channels or paths
3. Report any changes in the state of the paths up the stack
4. Perform actions required to fail over storage between controllers on the array

VMware vSphere includes a variety of generic plugins for storage arrays. I’ve identified the following:

• VMW_SATP_LOCAL – Local SATP for direct-attached devices
• VMW_SATP_DEFAULT_AA – Generic for active/active arrays
• VMW_SATP_DEFAULT_AP – Generic for active/passive arrays
• VMW_SATP_ALUA – Asymmetric Logical Unit Access-compliant arrays

Although I have sometimes seen other SATP plug-ins mentioned, the following plug-ins are all that are listed in the VMware ESX Hardware Compatibility List.

• VMW_SATP_LSI – LSI/NetApp arrays from Dell, HDS, IBM, Oracle, SGI
• VMW_SATP_SVC – IBM SVC-based systems (SVC, V7000, Actifio)
• VMW_SATP_CX – EMC/Dell CLARiiON  and Celerra (also VMW_SATP_ALUA_CX)
• VMW_SATP_SYMM – EMC Symmetrix DMX-3/DMX-4/VMAX, Invista
• VMW_SATP_INV – EMC Invista and VPLEX
• VMW_SATP_EQL – Dell EqualLogic systems

EMC PowerPath and HDS HDLM also support a variety of storage arrays, but I would classify these as full MPP replacements for PSA, rather than SATP plug-ins.

You can see which SATP plug-ins are available using the following esxcli command:

esxcli nmp satp list

Path selection plug-in (PSP) List

In contrast to the diversity of VAAI and SATP plug-ins, the universe of path selection plug-ins is fairly small. Most storage arrays are supported with either Most Recently Used (MRU) or Fixed path selection approaches. Many also support Round Robin (RR) path selection. The only vendor with a specific PSP that is not also part of a full MPP (like EMC PowerPath or HDS HDLM) is Dell, which offers a special routed path selection plug-in for the EqualLogic iSCSI arrays.

• VMW_PSP_MRU – Most-Recently Used (MRU) – Supports hundreds of storage arrays
• VMW_PSP_FIXED – Fixed – Supports hundreds of storage arrays
• VMW_PSP_RR – Round-Robin – Supports dozens of storage arrays
• DELL_PSP_EQL_ROUTED – Dell EqualLogic iSCSI arrays

As mentioned, EMC PowerPath also offers path selection as a plug-in in addition to the full MPP stack. Many other vendors offer unique path selection plug-ins, over 100 in total, but these are not specifically called out in the VMware HCL apart from their existence. I would love to learn more about them, however.

You can see which SATP plug-ins are available using the following esxcli command:

esxcli nmp psp list

Tell Me More About PSA!

As mentioned above, I am by no means an expert in VMware Pluggable Storage Architecture. Rather, I am interested in learning more and passing on this knowledge to others. I welcome your comments and feedback, and especially your corrections to the information presented here. I will try to keep this page updated as new versions of vSphere are introduced and as I learn more about this technology. Thank you for your help and understanding!

Further Reading:

• What’s New in VMware vSphere 4: Storage
• Next-Generation ESX Storage: A Pluggable Core Storage Architecture
• Storage Changes in the VMware vSphere 4 Family
• Pluggable Storage Architecture, exploring the next version of ESX/vCenter
• Vmware vSphere 4.1 PSA (Pluggable Storage Architecture) Understanding
• Explain the Pluggable Storage Architecture (PSA) Layout

© sfoskett for Stephen Foskett, Pack Rat, 2011. | VMware PSP and SATP in Plain English
This post was categorized as Enterprise storage, Everything, Virtual Storage. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

SME storage arrays like the AberSAN Z-Series pack solid enterprise-class features

It’s never been a better time to be in the market for enterprise storage products, with many excellent options available at affordable prices. But the market can be confusing for the uninitiated, with a variety of network options and capabilities. Even those of us “in the know” about enterprise storage are sometimes surprised by the offerings and companies in this space! So when Jerome Wendt from DCIG approached me to collect data for a market overview and buyer’s guide, I was excited. It was my big chance to really get to know these products!

Dividing the Storage Market

An enterprise storage array is a strategic investment, with prices often reaching into the hundreds of thousands of dollars. But there are smaller arrays as well, and these aren’t just stripped-down cheapies: Midrange storage arrays are emerging as serious challengers for the high-end enterprise arrays, and even the small array segment is improving. Long the home of simple RAID systems, small enterprise storage arrays are now full-featured systems with advanced integration and features.

This is the segment of the storage market I was eager to take a look at: The small-enterprise arrays, ranging in price from $5,000 to $30,000 and capacity from a few terabytes to over one hundred (with expansion shelves). To keep it sensible, I eliminated systems with less than 8 hard disk drives (since they really couldn’t deliver the IOPS needed for a production application) and those that can’t be shared using storage networking protocols like iSCSI, Fibre Channel, or NAS. This left a great cross-section, including small systems from big players like EMC, NetApp, Dell, and HP as well as smaller companies like Drobo, Aberdeen, StoneFly, and Promise.

Little Systems With Big Capabilities

Gathering data for this buyer’s guide entailed creating a master list of all the features any of the systems might have and then filling in the blanks with vendor input and public information. Although I felt that my list of questions was comprehensive, I was amazed to find that the available systems exceeded even my expectations. For example, I was surprised to see 10 GbE support in more than 1/3 of the arrays I looked at, yet only 1/4 of them included NAS protocols. Amazingly, every NAS system also supported iSCSI, meaning unified storage dominates the SME array market!

DCIG likes to rank the systems listed in their buyer’s guides, so it was left to me to come up with weights and scores. I decided to score each feature on a 1 to 5 scale based on my own expectations: A feature I was surprised by (more than 4 GB of cache, for example) got a 5 while one that was conspicuously absent (scalability beyond internal drives) got a 1. An average system would get all 3′s, but there really were no average systems!

I then weighted the features based on how relevant they are to small enterprise IT. Features like scalability (drives), data protection (RAID 6), and reliability (redundant power supplies and dual controllers) received more weight than less-critical things like ROHS compliance and even multi-pathing.

After all this work, we boiled down the scores to categories (controller, capacity, support, etc) and finally a single overall score. I was surprised at the results, really. I hadn’t spent much time with Aberdeen or StoneFly prior to this experience, but came away impressed by their products. The strong showing of EMC’s new VNXe and NetApp’s FAS2020 was as I had expected, but the D-Link DSN-5110, Dell’s PowerVault MD3200i, HP’s P2000, and the Promise and Winchester arrays caught me off-guard. These really-capable systems at low prices and should be on buyers’ short-lists!

Stephen’s Stance

I tried to put myself in the place of the buyer evaluating these systems, but of course every buyer will have his own priorities. Perhaps in the future we will make the raw data available so they can make their own ranking and determine which array works best for them. I will work with DCIG to make this happen, since the feature-focused rankings used this time aren’t right for everyone.

I have been working on this guide for almost a year now, and think the finished product really shines. I hope it will help buyers come up with a list of products to consider, and also that it helps the smaller vendors get some attention in this crowded space. Due to their strong showing, Aberdeen has licensed the guide and made it available for free (after registration) on their web site.

Download the guide and see for yourself!

Note that Aberdeen did not “sponsor” the creation of the guide: DCIG paid Foskett Services to create it with no sponsor in mind, I alone gathered the data, and our ranking and editing was finished before Aberdeen or any other vendor got involved financially. No one was more surprised than me by the strong showing of their AberSAN storage system, though the fact that it features Nexenta’s excellent software piqued my interest. Some have criticized DCIG’s guides and methodology in the past, but I personally stand behind this effort. I will happily answer any questions – just leave a comment here!

© sfoskett for Stephen Foskett, Pack Rat, 2011. | Which Small Enterprise Storage Arrays Are Worth Considering?
This post was categorized as Enterprise storage, Everything, Personal, Virtual Storage. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.