Reflecting on a long year
It’s been a busy 2014. We’ve achieved a lot in the ROC, and have been intensely engaged in developing new services for our user and technical community. Beyond maintaining standard grid services, and keeping existing sites operational, we’ve tried to identify and deploy new sites in the region. Beyond standard grid services for computing and data storage, we’ve branched out into other services designed at reducing the barriers to entry for new users and application developers. Work on identity providers, science gateways, and bridges from the web into the grid have been some of the highlights of this year. We’ve spent a lot of time trying to design and implement a platform that will make it not only easy to port, integrate and deploy new applications, but also keep track of their usage and impact by scientific communities.
All in all, it’s been a good year. But you’d never say so, looking at our websites !
So much effort has gone into this that we’ve neglected somewhat our ‘presence’ on the web. As with so many things, this was due to a lack of time necessary to take a concentrated, hard look at what we were doing and have enough breathing room to think clearly and start anew. The Christmas period is a traditionally slow time, both for users and technical staff, as well as our administrative and managerial cohorts. Thanks to these few days, we’ve been able to redesign this here website. We’re about ready for prime time, when we’ll tag the repo with a new version. Here’s a sneak peak at what’s coming.
The template we’re using is based on bootstrap. The site is made with Jekyll, with some help from fontawesome, jquery and the github API itself. We took some time to learn a bit better what these frameworks were good for and made some significant changes to the templates and code, which has really helped to rationalise the content. Apart from some layout, font and other design changes (because we’re sensitive like that !), here’s an outline of the rationale behind some of the changes.
A meaningful front page
This was just a blog. Necessary, yes, but not sufficient, for a technical community serving research communities. We’ve added a nice new nav bar to the front page, which we’re slowly fleshing out with content. This should give the casual visitor a better idea of what we do, who we are and what’s going on. You’re still going to get surprised by the monkey from time to time, but hey, we’re working on it .
Easy links for you fine folk
Different folks for different folks, right ? We want to make this a friendly place not just for the hardcore operators, but also for the wider community of users and expecially developers which may be directed here - or who knows, just stumble across us by accident ! It’s no secret we’re, like invested in github. Like… heavily invested… So, we decided to adopt some of the octocats to represent them.
Services for Researchers
Make the grid better with us
How we roll
Our friendly robots
For researchers, we’ll include some description of the higher interfaces to the infrastructure, such as science gateways, etc, along with links on where to sign up, ask for help and generally use what we’ve got. Some description of what the grid is good for, when it comes to producing and curating scientific results will of course go there as well.
No party is fun if you’re the only one there. We want the ROC to be a fertile place for ideas and innovation, and as such, it needs to be developer friendly. This was one of the main reasons behind moving most of our kit to github, but beyond that we need to provide a better view of where the ‘front doors’ are. Descriptions of middleware API’s, service endpoints etc… you know, just like every other service out there.
Beyond that, we’re always looking for project ideas and interested hackers to work on them. This includes University students and postgraduate projects, but also the wider community of people who are out there on the web. If you have and idea or a wishlist, be it a new application or a whole new platform, we want to know about it and see whether we can deploy it together.
These are the godlike beings behind the services. Invisible, all-poweful and benign, they bend over backwards - between rescuing folk in the wild, preparing haggis, snowboarding in the mountains, being good parents, biologists, philosophers, etc - to ensure that services are up. The operators need and deserve better visibility and access to the tools which make us a team. The team spans more countries and institutes than you can shake a stick at, and it’s this very diversity that forms our strength. However, a common point of reference is crucial to ensuring that we’re all speaking the same language on the operations centre, regardless of that which we may speak at home. Which, of course, varies widely from site to site…
Finally, our beloved bots. These are newcomers to the scene and are mostly still under development, but we are betting heavily on automation. From site deployment to application porting, there are many things which the likes of Puppet, Jenkins etc just do better than we cruddy humans. They too deserve the bit of prime time.
Highlighting the good work
I, for one, am tired of writing reports and emails that few people read. If something is worth working on, in this game, it’s worth talking about - and the more people read it, the better. We’ve conducted a few operations campaigns during the course of this year, which required a consistent, coordinated effort by most or all of the sites. Furthermore, we’ve got a few projects on the go, which require deeper and more concentrated development. These need to be better highlighted and communicated so that the people who work so hard on them can take the credit due to them. They also help publicise and promote areas where contributions or collaboration would be welcome (and soon, a place where people can suggest new projects for us to adopt that they might be working on). Better access to information will mean better collaborations.
Quick and dirty dashboard
Finally, we’re working on a small, simple dashboard to give a snapshot of the state of our core services, but also those not covered by SAM. This is not meant to be a reimplementation of the Operations Portal, or the Virtual Organisation portal. It will give simple info about the state of the services which we use which are not included therein - the core services, the dev cluster, various websites etc that are outside of the realm of the grid, but still important to daily life of the ops team and relying communities.
Watch this space - or help out !
Although most of the layout and structure has been done, some content and logic needs to be implemented. This is still a work in progress of course, so check back (or subscribe to the rss - or better yet, the watch the repo. If you like what you see and want to give us a hand, you know what to do - fork and send us a pull request. We’ll be ever so glad !
Till next year !
So, here’s to a great and industrious year gone by. We’ve achieved much and are in a very good position to build on the work done. I’d like to thank each one of the operations team, from all over Africa and the Arabian peninsula, for their hard work and dedication to a pan-African shared computational infrastructure.
Let’s heave one last sigh of relief that the year is coming to a close, and hit 2015 right in the sweet spot when it rolls around in a few days.
Bruce Becker: coordinator, Africa-Arabia Regional Operations Centre