Announcing the Cloud Performance Summit

We’re introducing a new event to Cloud Connect this year, and it’s an indication of how much utility computing has matured since last year’s inaugural event.  Here are some thoughts on the Cloud Performance Summit, and why performance may be this year’s hot topic for on-demand computing.

In the early stages of any industry, the discussions focus on the “why” and “what.” Clouds are no different: we wanted to know what clouds were—with the inevitable debate over taxonomy and definition—and we hunted for reasons to embrace them, or to refuse them, depending on our own agendas.

But by now, most enterprise IT professionals have accepted that cloud technology is inevitable, and that third-party cloud providers deserve a place in their toolbox. Put another way, we’ve moved from tender embraces and heated arguments to the dispassionate world of the prenup. We want to know, can clouds deliver, and if they can’t, what can we do about it?

Performance is a tough subject. For one thing, cloud providers offer a shared resource. It’s the basis of their economic value proposition. And a shared resource means things like oversubscription, badly-behaved neighbors, and having to fight for service quality.

But it’s not just about sharing computing resources with others. For decades, IT has worked with a simple equation, namely, that the performance of a system is a function of how many people use it, and how much capacity it has. Roughly speaking, more users means a slower application, and more computers means a faster one.

Clouds offer capacity on demand. They’re elastic. Which means that in the demand/capacity equation, capacity is effectively limitless. If you want things to go faster, you can pay for additional capacity. And that’s why performance matters: it’s directly tied to your costs.

Consider air conditioning. With your own power, there’s a limit to how much you can cool a house. If you want it colder, you don’t have enough electricity to run your appliances; if you add a bigger generator, you can cool it more. But once you’re hooked up to an electrical grid, you can cool the house far more—and your bill will show that. With clouds, it’s not cooling, it’s performance.

Badly written code costs money, too, when you’re paying by compute cycle. Amazon’s Cloudfront CDN is forcing other application accelerators to offer pay-as-you-go pricing, which means more and more of the performance problem is now a billable cost.

At Interop New York, a panel of performance experts concluded that performance may in fact be a bigger problem than security—after all, there are security certifications on which customers can rely, but there’s precious little guidance when it comes to outages and latency. A Queen Mary University study concluded that the vast majority of cloud providers offer no guarantees in their terms of service, and if they do, then compensation is limited to a refund of service costs.

Making things even worse is the complexity of cloud deployments, which often involve many providers and components, and are harder to diagnose and instrument than in-house, centralized applications.

So we’re really excited about the summit. It’s bringing together vendors, end users, and performance experts in a relatively informal, open format to discuss some of these hard issues. It’s the first time we’re running it, but we’re already certain it won’t be the last.

The ROI of clouds, from IGT09

I’m at IGT09 this week, put together by the energetic Avner Algom. Yesterday, I gave a presentation on the ROI of cloud computing (including some data that IDC and Peter Van Eijk were nice enough to let me use.) We started off with a panel on enterprise cloud use, with four panelists from very different backgrounds:

  • Steve Rubinow, EVP and CIO of NYSE Euronext
  • Yosi Shneck, CIO, Israeli Light Company
  • Eyal Waldman, Co-founder & CEO, Mellanox
  • Liam Lynch, Chief Security Strategist, eBay

Here are the slides, including panel questions.

Getting ready for the Future of Clouds with Dr. Bob Marcus later today. Should be fun; he has a lot more useful information than I do to share with the crowd, so I’ll try to make vague, generic comments that can’t be proven instead.

Conclusions from the Hybrid Cloud panel

Lots of good conclusions from the Hybrid Cloud panel at Interop; too many to see on one slide, so here they are.

  • Hybrid isn’t one app in two places; it’s internal and external apps talking to one another
  • Migrating for new apps is easy; for already deployed ones, it’s much harder
  • In the new world, the developers are the admins and ops toolsets are changing
  • The 2010 platform will be
    • Infrastructure-aware; parallel; split between dev and ops
    • Not really PaaS; but not IaaS either
    • Runs both in-house and externally
  • Increased focus on making it easy for developers to transition to on-demand environments
  • Portability becomes a bigger concern (in/out and between clouds)
  • Where enterprises will initially embrace it:
    • Collaboration, messaging, things “just above” infrastructure
    • Areas that don’t add strategic value
    • Leverage utility model of what’s there now (apps with inherent burstability)
  • Ideological battle in infrastructure
    • Bottoms-up focus on primitives (storage, queue, compute); we build things from easy-to-connect, RESTful functions
    • Top-down modelled approach, which we reduce down to the underlying patterns and can generate code from them (policies, etc.)
  • Growth of 2 kinds of technologies
    • That make this easier for developers (Ruby on Rails) ➜ This will win
    • That help to migrate legacy systems into cloud-compatible containers
  • Enterprises about 5-7 years behind consumer/public Internet (Web tech, Hadoop, enterprise mashups)
  • Let us not forget: All big web businesses use a ton of Oracle
  • Huge $ to be made solving enterprise migration: Discrete components or specific apps
  • Standardization is 15% of the problem; standards bodies are still arguing taxonomies
  • Internet standards are built on rough consensus and running code; whoever produces a useful product that is available to many people quickly will win

Why elasticity, performance, and analytics will change how Webops is judged

I got to Velocity this morning, and Jesse asked me if I wanted to get on stage for five minutes to talk before lunch. Given that I’m doing a session in the afternoon called What The Rest Of Your Company Knows About Your Website, I figured I should make something new.

One of the things that’s abundantly clear — echoed in presentations from Shopzilla, Google, and many other excellent speakers — is that performance matters. It affects your conversion rates; it even changes your Search Engine Marketing ranking (which was news to me.)

Continue reading “Why elasticity, performance, and analytics will change how Webops is judged”

Interop Cloud Camp 2009

One of the things I’m most excited about at Interop this year is CloudCamp. Dave Nielsen, Sam Charrington, and some of the other folks behind the CloudCamp events are bringing it to Sin City!

These fast-based, organized-on-the-spot events are always entertaining. And with over fifty cloud computing experts in town for the Enterprise Cloud Summit, it’s going to be a great discussion.

The event is open to attendees of Interop, so just register for a free expo pass and you’re in. You can also come if you’re attending any of the Interop content, such as the general conference or the workshops. And you can of course sign up directly with CloudCamp.

Our hats are way, way off to the folks at Interop for providing us the floorspace and some food and drink to make this event possible. It’s a great contribution to the cloud computing community.

Enterprise Cloud Summit '09

Enterprise Cloud Summit is getting ready to launch. We’ve listed some of the speakers, panelists, and participants for the event, and I’m pleased to say they include some of the most interesting thinkers in cloud computing: Werner Vogels from Amazon, Ben Black from Opscode, Mike Repass from Google, Lew Moorman from Rackspace, and lots of others. We also have folks from Forrester and Booz Allen Hamilton joining us.

One of the things I’m most excited about, however, is the demos. Dan Koffler of Syntenic is coordinating six live demos over the two days, showing how to build, run, and scale cloud-based applications. It’s always dangerous to do a live demo at an event, so we figured we’d get on the bleeding edge and run six of them, back to back.

ECS is a paid event that’s happening in conjunction with Interop Las Vegas. Since Bitcurrent is running the event for the folks at Interop, they gave us a $100 discount that you can use by following this link this link.

The Enterprise Cloud (part 1.5)

There was a question from the audience: “Gee, Steve, what’s the difference between what you’re proposing and straight up virtualization?”

Good question. Glad you asked. Good enough question in fact to insert part 1.5 inbetween parts 1 and 2.

The definition of cloud computing remains nebulous at best. We’re entering a phase where everything is claiming to be a cloud — if you offer something hosted, it’s a cloud. By such a loose definition, the tech biz has been selling clouds since we’ve been renting mainframe time. To offer a little contrast, Amazon EC2 is a huge cluster of virtual machines that you rent a-la cheap dedicated virtual servers. Continue reading “The Enterprise Cloud (part 1.5)”

Why I like my data near my logic

I use an email client on my iMac. When it can’t get to the server, it still works. But sometimes, on a slow day with an unreliable network like the one I’m on right now, I don’t realize that I have mail waiting for me. The disconnect between by client-side logic and the server-side data camouflages the fact that the network isn’t working.

By contrast, when I use GMail’s web interface to read my mail, I know when I have new messages. Because Google controls the processing (on its servers) and data (right next to them) the two are connected. No camouflage there: If the network sucks, I know it.
Continue reading “Why I like my data near my logic”

Cloud for the SMB – That ship has sailed…

I have been talking to a lot of prospective cloud management start-ups lately, and a theme I am hearing repeatedly is that SMB is the great untapped opportunity. Most are hoping to be the RightScale for the SMB market by providing them with super simple web-based interfaces to clouds like EC2. What I’ve been telling them, is that unless they define SMB the way IBM does (when I was last working at IBM in 2005, eBay was classified as SMB), Cloud for SMB is a ship that has already sailed… Continue reading “Cloud for the SMB – That ship has sailed…”