James Governor's Monkchips

CloudCamp London: the inauguration

Share via Twitter Share via Facebook Share via Linkedin Share via Reddit

 

CloudCamp London rocked: around 250 people showed up, and the applause for the speakers was surprisingly generous from a geek UK audience; its clear there is a hunger for information in the space. The format wasn’t perfect, but it was still a very good effort, even if there were no dead dragons lying bleeding afterwards.  

Reuven Cohen, CEO of enomaly, the guy behind the original CloudCamp San Francisco, kicked off proceedings by saying that the event series was going global- in the next few months we’ll see New York, Chicago, Boston and Paris all hold their own. Like BarCamp this isn’t a top down phenomenon– anyone can get involved, as a speaker, sponsor, attendee, or whatever. Check out www.cloudcamp.com for more info. We’re working on governance ideas now….

The first speaker was Simon Wardley, otherwise known as the *AAS Master. Unlike some of the other speakers he actually stuck to the Lightning Talk format’s 10 minute limit. Key phrase: “Yesterday’s hot stuff becomes tomorrow’s boredom“. His presentation was the usual mashup of ducks and horror movie stills. The essential argument- unless open source standards emerge, cloud will just be another round of industry lock in. One take on that argument is that customers always vote with their feet, and they tend vote for something somewhat proprietary – see Salesforce APEX and iPhone apps for example. Experience always comes before open. Even supposed open standards dorks these days are rushing headlong into the walled garden of gorgeousness we like call Apple Computer.

Next up came Adil Mohammed, co-founder of entrip, with his take on why the cloud is perfect for startups. Adil made some good points, though others felt like motherhood and apple pie – startups want reliability (try telling that to Twitter, which almost seems to benefit from downtime).

“Cloud Computing has leveled the playing field with enterprises.”

I was also a bit worried by the numbers he pointed to for Animoto, an example of scale via cloud. People are really liking animoto, thought its not without problems. But according to Gojko, Mohammed pointed out that:

Animoto, which grew from 25000 users to 250000 users in three days, scaling from 50 to 4000 servers in that time and growing at peak 20000 users per hour. The cloud deployment made it possible to do that, since growing that fast on a dedicated infrastructure would simply be impossible even if already purchased the hardware.

Wait. wait. wait… just a second here. According to my calculations that’s a new server for every 63 users. Cloud or on premises that’s not exactly an impressive scalability number.  I guess I need to know more.

Other speakers were not as generous to the community as they should have been, frankly. Brevity for the win. Given that fact the middle part of the evening felt more like marketing – particularly the Gigaspaces pitch, which was the same ol same ol’ virtual tiers presentation with one or two cloud mentions for good mention.

I thought Martin Buhr of Amazon did OK, though, and announced European hosting buildout (which should be good for network bandwidth costs and data protection legislation).

I think Will Fellows of the 451 Group did an excellent job of putting foward a cloud taxonomy. Really I was very impressed with the work he has put in. I would strongly advise you to ping him and ask for a free copy of the report his talk was based on (the offered one to cloudcamp attendees.

Last up was Alan Williamson. summarised eloquently here:

“His main message was that with cloud infrastructures problems don’t magically go away, they just shift. You don’t have scalability or storage problems any more, but you need to constantly monitor the cloud and your application in it. Alan pointed out examples when Amazon’s cloud failed and their applications got cut off from the Internet. As a solution, he proposed deploying the application on more than one cloud so that you have resilience. This requires writing the application in a way that can be easily ported to different providers, which in itself might be a challenge. One idea that was really striking was their analysis of getting off the cloud to a dedicated infrastructure again — apparently it would take them about three weeks of full-bandwidth transfer to download the data that they have in the cloud, making it virtually impossible to go back.”

Nice- so much for “freedom to leave“. The service might support it, but with massive data sets, portability ain’t so easy… Mi compadri Stephen O’Grady recently posted some good thoughts on Cloud Standards but its also worth considering the physical limits of data portability (we might be talking about flowing a terabyte of data, not just an email address). To often we assume everything on the web is instantaneous. We’re talking about the Physics, rather than the Economics of Data Portability. Data volumes will certainly be a key challenge for data portability, which is one reason my money is on the Synchronised Web. 

Well perhaps not everything is instantaneous: as Will Fellows said, he was talking to a middleware vendor who said he wasn’t losing sales to Cloud Computing… but it was elongating the sales cycle. Not sure how that fits into 15 Ways To Tell Its Not Cloud Computing… I know there were quite a few suits at CloudCamp. 😉

Final Thoughts:

The event was a success, but the key takeaway is that the format needs more user participation. The sponsorships were frankly not expensive enough to justify vendors using them as marketing spiels. CloudCamp needs to be more about the community, as Alexis Richardson, the point man for the London event, explains better than I could:

“Which brings me on to the ‘Open Spaces’ aspect of the event and the lightning talks. Putting it about as politely as possible this is the area ‘most in need of improvement’ for next time. We’d set a limit of ten minutes per talk which of course everyone effortlessly broke. This was entirely my fault for not bringing a stop watch and baseball bat. Next time, all lightning talks will be ruthlessly limited – possibly using an Ignite format. Admittedly some speakers were so engaging that the time flew by and the questions could have gone on all night. But – that’s why we brought beer and pizza.
Secondly this was the first Open Spaces event for many people who came. I must confess to having been nervous about this beforehand, and on the web site had asked people to propose talks or topics. About twelve people were enthusiastic or sympathetic enough to offer talks. I figured this would be enough to motivate others to step up from the audience but, aside from one or two fluent folks such as Mark Masterson, it just did not happen. Argh.
At the SF event there were many talks and many people who like to talk. Was London CloudCamp making people laconic, or worse yet, shy? Based on the fantastic atmosphere throughout, and the great conversations people had till late in the evening, I don’t think so. I think we can make the format much better. One idea is to have more smaller rooms and start with an hour or two of Open Spaces unconference, then mingle over beers, and have a few really rapid fire talks after that. Please do comment on this blog if you have suggestions!”

Special thanks to Skills Matter for being the operations people behind CloudCamp London. I believe the phrase is “we couldn’t have done it without them.”

 

picture credit Chris Purrington. He says “all rights reserved” but I am sure he’ll consider this fair use.

9 comments

  1. The work that Animoto does on EC2 (taking images and music and dynamically producing video) must be extremely CPU intensive. According to something I saw, it takes around 5 to 10 minutes to generate a video, using up most of the capacity of a small instance during that period. So, I’m not sure if they have the revenues to cover the cost of them, but I can see how they could get up to that many instances. Since their usage pattern is pretty different, I wouldn’t use Animoto as a gauge of how many concurrent users an EC2 instance can handle for a typical website.

  2. Right on Matt. As you say video rendering is an incredibly CPU intensive process.

    I hope it didn’t appear that I was talking about the limitations of anything other than a specific architecture. Of course animoto is not representative of all web architectures.

  3. Hi James, thanks for the reply. I was specifically responding to this paragraph: “… that’s a new server for every 63 users. Cloud or on premises that’s not exactly an impressive scalability number. I guess I need to know more.”

    I was just meaning to say that it seems impressive that EC2 could handle the load gracefully (as far as I know), aside from whether Animoto is a well-designed or practical architecture. I guess we were talking about different things. 🙂

    As a side note, it seems like all of the prominent examples of scaling on EC2 involve batch-oriented, asynchronous processes (for which the SQS/S3/EC2 combination seems well suited). It would be interesting to hear about a large scale interactive app running on EC2.

  4. Wait. wait. wait… just a second here. According to my calculations that’s a new server for every 63 users. Cloud or on premises that’s not exactly an impressive scalability number. I guess I need to know more.

    It wasn’t just processing of new user registrations. They added a Facebook application that allowed new users to create their first Animoto video. That’s why they needed all those servers. Also it was 25k to 700k. Here is a podcast w/Brad Jefferson the CEO of Animoto if you want the details.

    Cloud Cafe Podcast – Animoto

  5. hey John nice to see you, but really, did you read the comments to help with context? I clearly never said all the servers were just for handling new user registrations.

    I think Matt does a great job of helping to clarify.

  6. […] was actually James Governor’s Segway (founder of RedMonk – his write up of the event here: CloudCamp London: the inauguration). However, I didn’t manage to persuade James to part with it, but he did get me thinking […]

  7. […] a enterprise level point of view what happens if for whatever reason you want to leave the cloud. James Governor points to this problem when he quotes Alan Williamson in a […]

Leave a Reply

Your email address will not be published. Required fields are marked *