What they’re doing is pretty straight forward: they’re monitoring the few metrics that are available for each service, from multiple points globally. What they don’t do is let each person monitor from their perspective. Instead, CloudStatus provides the “what everyone is seeing” 3rd party perspective.
Clearly, narrowing down to each individual “user” is something that someone like Hyperic would do well to productize rather than simply give away for free. Indeed, as the use of cloud computing for an extension of data centers evolves, you could see a stronger linking between Hyperic’s main product, HQ and something like Cloud Status.
What’d you want to do is link performance monitoring from a users desktop, through the company network, up through some servers, connecting to the cloud, and then round-trip back to that user. You know, the end-to-end fantasy all the ESM old-timers get excited about.
Starting to Establish Trust
While Cloud Status doesn’t provide that, it starts the ball rolling with two important things:
Providing visibility into Amazon EC2 - there aren’t a whole hell of a lotta services that simply do monitoring of *aaSes. Sure, you can ping things all day long - check for basic uptime - but the metrics beyond uptime that Cloud Status is monitoring points towards much more useful monitoring. For example, monitoring how long it takes to deploy a new instance in EC2.
Historical tracking - while Cloud Status doesn’t do this extremely well (you’re restricted to the date range selected in each graph), clearly, Hyperic is building up a historic record on AWS’s uptime. Getting a handle on the performance trends for stuff in the cloud is key for people who want to build enough trust in the cloud to actually use it.
Indeed, this last point marks a good mile-stone that Cloud Status pushes us towards in this grab-aaS infrastructure world we’re hurdling towards: the tracking and visibility into services and infrastructure our of control that helps build trust in those services and infrastructure.
There’s Always More
Now, there’s much left to be done along these lines. Our friend, the cloud to everyone’s silver-lining has a laundry basket of vitriol along these lines. There’s a sort of curse of The Age of Beta going on here: either you release early, release often and get criticized for not doing enough, or you wait for the full 1.0 release and risk being late to market (with the needs changing) and being criticized for being slow.
Hyperic knows there’s room, and deep-desire, for more, for sure, and is of course “working on it.” Indeed, as I understand it, moving this rock up the hill is a large part of what Mike Olson is doing with Hyperic, which is encouraging as use RedMonks are unabashed Olson fans (who isn’t, really?).
On another, sort of insider-baseball note, it’s interesting to see that Hyperic’s re-position from general, replace the Big 4 ESM to monitoring and management for web based concerns is falling into place with all this cloud computing glee.
Monitoring, as it were, Hyperic and other’s path along those lines is what’s key to the ideas beyond Cloud Status. What we want is to get an arms race going between Hyperic and others to push monitoring (and management!) forward so that once all these clouds pop up (as they’re fixin’ to do) we’ll actually have the trust that comes with transparency to use them. Otherwise, it’s getting stuck in the swamp of our own data-centers, which is good work if you can get it, but a bit too much centralization for the wider trickling down all this cloud stuff could get us to.
Disclaimer: Hyperic is a client, and Javier also bought me several tasty Blue Moons recently.
Previous Topic
