Impact of Cloud

Workshop on cloud computing.

Cloud is inescapable. Huge growth. Many vendors now only offering cloud solutions.
Benefits, increasingly towards time to market, faster access to infrastructure.

Size of device has gone down, amount of data has increased. We also own many more cloud enabled devices. Can set up old devices as cloud devices, eg as webcams. As we get to Internet of things, even smaller, more devices, much more data. Struggling now. Will get worse

Use cloud at home, fitbit, Spotify, app stores, music etc

At work, Dropbox, Google apps, azure, salesforce

Our expectations of our relationships with organisations have changed. We're used to accessing services anywhere on any device. If we don't like something, will just download another app. Having the services we need. We expect companies to look after our data securely.

So, organisations are changing. Multiplatform provision, agile development, rapid release. Security around applications rather than perimeter. Proactive monitoring of social media. IT governance. Making workplace employee experience as slick as consumer experience.

The IT function is changing. Change in control, consumers have choice. Change from data to information. Changing the way we recruit. We have information security mangers. UI/UX teams, more API development platforms. More contract managers. More business partners. More process improvers.

Also, our data is in many more different places. Do we know where it is? Mainly no :-)
Vendor maturity is a challenge. Some build up business, then fail. Vendor error also an issue, gives a different sort of risk profile. Failures can cause huge impact on our services. We are dependent on them, services are out of our control.

Early stage impacts of cloud include access to massive compute and storage resources, access to content and learning, access to new platforms and social collaborations. But, what are future impacts going to be?

Despite all the hype about cloud computing, we have mainly kept doing what we've always done with new applications and new titles, the real impact of cloud is only just starting to be visible.
IaaS, PaaS, SaaS is where we are.

Where we're going is machine learning as a service (MLaaS) and Business process as a service (BPaaS) this is where new impacts will be.

Cloud only works if you think it through end to end. So, if you put all your services in the cloud, they will still be down if your connection goes down. But, they won't be if authentication to those services is still on campus. Can't think where he got that example from!

Real impacts of cloud. You store your data in the cloud, your organisation loses the data when you leave. Your data is sold without your knowledge, you can't get insurance. Your data is shared for medical research purposes, and your life is saved by early intervention due to predictive analytics.
Internet of things, machine to machine interaction, there will be huge impacts.

 If the fridge won't speak to the oven, you may have to go out for dinner....

Posted with Blogsy

RUGIT - Security, Cloud and Innovation

Today I've been to a RUGIT (Russell Group IT Directors) meeting in London at Kings College. A great location by London Bridge with the Shard towering over it.

First up was a session from JISC on security issues, starting with the implications for us of the Counter terrorism and security act 2015. It says that we should have due regard of the need to prevent people for being drawn into terrorism. It also says that as a university, we should have particular regard for promoting free speech.
Guidance on implementing the act has been published and is available here, and it advise that it should not add large new burdens to institutions who are following current best practice. Most of guidance is around updating policies and processes, including acceptable use policy which is expected to mention the new statutory duty, but no wording is currently suggested.

There's no real technology implications, other than if an institution already filters harmful content, then you should consider adding this.
JISC is currently developing on line staff awareness training.

We then looked at some other security issues, around our complex environment which covers everything from providing home broadband to complex research environments.
We have very diverse security requirements in same organisation. Because of this, we've been dealing with issues around things like BYOD and incident response for more than 20 years.
But, because our environment is changing, and the smartphone in our pockets has more power than universities had 20 years ago.
New habits are now routine. Mobile working and the blurring of the life/work means that safe IT behaviour is no longer something you need to do at work. it's a life skill.
Security can't be done by IT alone. Our users no longer need our hardware and can change security zone at the click of a mouse.

The role of IT department is to look help our organisations adopt and choose a package of  behaviour, policy, and technology.

Think "work safe", not "stop unsafeness"

We also had a session from a layer, an expert in cloud  - he has a blog which I've a had a quick look at and it contains some interesting stuff - on cloud risks and how to manage them

The first was SLA oversell, where the sales pitch says the service will be 100% available, secure, unhackable, the best, fastest, cheapest etc. however, the SLA will contain phrases like "make  Reasonable efforts etc".
100% availability except, scheduled maintenance, planned maintenance, unscheduled maintenance, emergency, etc.

Another clause found in lots of cloud terms and conditions is the "As is" service.
Or, the service is as we provide it. It might work, it might not.
There's often other clauses excluding any warranties, with no guarantee over what they provide. No guarantee that data won't be lost. They're not liable to you for any losses, even data. All of the above are contained in a the Ts and Cs of a very big web services company who made their name selling books...

How do you manage it? Negotiate terms? But often can't with big corporates. 
Pay more, get better cloud? Pay for failover, redundancy etc?
Split between public, private and hybrid?

Issues around data compliance tax a lot of people, and we concluded that wherever your data is someone will be able to get to it
 Every country has a surveillance organisation, and some legal jurisdiction to get at data.

How do you manage this? Keep you data in your local data centre? Use a hybrid cloud? Encryption, tokenisation? Or just do a robust assessment of the risks

The other risk with cloud services is Disaster recovery and insolvency. First have gone bust, and administrators have demanded large sums of money from customers to get their data back.  You need to plan for the worst and have a DR strategy

finally we had a session on innovation management from Oxford and Birmingham. Both have implemented solutions similar to a Ideascale for generating, capturing and scoring ideas. Done in slightly different ways in the two organisations, but with similar results. Innovative ideas are sought in a campaign from staff and students, and are voted on, and the assessed by a panel. The most successful are funded. in some cases, the staff or student originator works on the project, in some they are developed by the It at department.

Very good day. Always good to meet the others and share ideas and issues

Cloud, crowd and outsourcing will eat your lunch

Presentation from Cornell University, Enterprise cloud strategists.

Disruptive changes are happening in IT sourcing.

Cloud computing is the obvious one. We can't compete with it. Commoditisation, scale etc. No brainer. Everyone should be doing it for commodity services.

Crowdsourcing will reshape IT staffing. Gartner have predicted that in 15 years the predominant employment relationship for IT staff will be freelance/contract based. Will have to reach outside of our community for some skill sets.
Cornell have done a crowdsourcing pilot with TopCoder. They have a global, talent pool approaching 1m. It's competition based, and Cornell only pay for the best solution. Projects broken down into very small projects which encourages hyper specialisation. The individual solutions then Integrated by specialist integrators.

Outsourcing is rising. Big companies like Accencture, Capgemini offer packages for outsourcing whole services

Disruptive innovation is accelerating. Time between disruptive innovations is decreasing. Foundations are being shaken everywhere. Digital experiences are replacing people experiences. Uber vs yellow cabs. Airbnb vs hotels. Travelocity vs travel agents.

Cloud vendors have the ability to disrupt us, can go to end users and bypass IT departments. They sell above IT directly to business units, and below IT directly to end users. We are no longer necessarily the providers of services.
We want to wrap our services around cloud, but consumers want to just buy products.

Enterprise IT roles are changing.
Oliver Marks from ZDNet quote. "Cloud companies are cost effectively emancipating enterprises from the tyranny of IT, solving lots of problems with tools that are a pleasure to use."

So, campus goes shopping, but the problems will still be ours! We need to change our relationship with our customers. Bridge the gulf between what our current structure/staffing was built to do and what is required of us in the era of post enterprise IT

We need to be a business asset by becoming the following:
Expert advisor for disruptive change
Navigator of procedural barriers
Innovation hunter
Time to market experts
Bridger of gaps (integration, architecture, security etc)
Value added reseller of cloud services, ie wrap our support etc round cloud services
Be or support an IT VMO (Vendor Management Office)
Refocus ourselves on supporting the core mission of teaching and learning and research. Running IT is not the core mission of the University.

Enterprise IT must become the strategic advisors to our customers and our vendors.

Don't be a mere operational function, a follower in a era of disruption and commoditisation
Be a change leader, a business asset, aligned with the core mission.



- Posted using BlogPress from my iPad

Changing Landscapes

Yesterday I gave the opening presentation at the UCISA Changing Landscapes event organised by the Staff Development Group and held here in Sheffield. My job was really to set the scene for the day, and I decided to outline what the "Changing Landscape" was from my perspective as an IT Director, the way these changes are affecting how we deliver services, and the impact on the skills needed by our staff and students.

As usual, I enjoyed giving the presentation, despite staying up late the night before writing it - it doesn't matter how much notice you give me (almost a year in this case), I'll still be finishing it off
right up to actually standing up and talking! Changing Landscapes was perhaps an appropriate title for a conference given the effects of the weather at the moment!  The main trends I covered were:

• Consumerisation
• Mobility
• Cloud
• Social Media

All are linked, and together with the rapid changes in technology we're seeing, are having a major impact on the way we deliver services.

To illustrate consumerisation of IT I like to find some gadgets that are around, or just being developed - a sensor in a babies nappy with tweets you when it needs changing, a football with a sensor inside to tell you how to improve your game, an internet enabled fridge which send recipes depending on what's in it to your internet enabled fridge. These all go to make up the Internet of Things I mentioned the other day, or the Internet of Useless Things as someone referred to it yesterday....

We had some interesting stats on mobile, where the number of students owning tablets or eReaders has gone up from 7% to 29% in one year. We are definitely entering the post PC era, and its expected that tablet sales will outstrip PC sales sometime this year. A recent Gartner prediction is that by 2017 there will be more words typed on glass than on keyboards. And the iPad was only released in April 2010.

 I like to showcase a bit of what we're doing here in Sheffield, so I showed how we use social media to interact with staff and students, using our Twitter feed and Facebook pages as examples where our aim is to have a conversation, and not use them merely to give out notifications. And we're looking to engage with innovative videos and infographics. No-one can forget our Save it Like a Hero video (which despite much criticism from within the department has been a huge success with the audience it was aimed at - students), and today we have a Valentine's theme to our tweets and posts - Fall in Love with Safe Computing. Sweet.

The talk was filmed, so I'll post a link when it's up, just in case anyone is interested in watching it.

Cloud Advisory Board

Today I was at the JANET Cloud Advisory Board. This was established when the JANET Brokerage was set up with £1.5m of UMF money and has acted as a Steering Group for the brokerage activities. Now the UMF money is coming to an end, and we were meeting to look at how we go forward with JANET activities, especially in the context of changes to the JISC.

We looked back at some of what the brokerage had achieved:
Frameworks are in place for data centre to cloud and telephony services
New frameworks are about to be announced for archive to tape and file synch and share
Two commercial arrnagements are in place  - Google and Microsoft
Financial X rays have been carried out in 18 organsiations

In the future we agreed that as a sector we need to  be more confident and assertive with suppliers, with a more commercial attitude to providing services. We also need to be ahead of the game more - speed is essential and as the digital world is moving fast we have to keep up. We must become thought leaders, and the inertia for innovation has to be banished!

We agreed that we need to work together - collaboration with ourselves, with JISC and Janet, with our suppliers and other organisations is vital. 

So, we will continue as an Advisory Board but maybe with a slightly different membership, and will act as a "critical friend" to JISC technologies which consisits mainly of JANET and Cloud services, advising on product offerings and services. We will also input into the different stages of a product life cycle including requirements gathering, testing etc. I'm looking forward to working with the new JISC, especially the new Director of Innovation,  on these developments.

During the day we had quite a lot of opportunity for discussion around cloud services, and one of the things I'm interested in is why more Universities, including us, aren't moving more into the cloud. Storage seems to be a favourite for many people, including many commercial organisations. Yet we, and I suspect other HEIs are spending vast amounts of money on on-site data storage. So, what are the barriers, why aren't we doing it? I asked the question on twitter and was directed to this article by Andy Powell which suggests that Universities think of themselves as special cases. Other reasons given included leveraging the already heavy investment we've made in on-site provision, cost, complex inter-dependencies with other systems, bandwidth and latency. All good things that need to be taken into account, but I think we really do have to look carefully at what we provide internally and what could be provided elsewhere, in a public, private or hybrid cloud. Cost comparisons need to look at the total cost of provision, including space, power, staff, and not just hardware and software. of course, moving stuff into the cloud has to have real benefits, in either service provision or cost efficiencies to make it worthwhile.

I've looked at clouds....

We had a session today from Tim Marshall, CEO of JANET on Cloud.

One of the first things he suggested that we should do,is make sure we know what's going on around us. Don't do things in a particular way just because we always have done.
A wonderful example is TVs coverage of tennis matches which is always for. The end of the court. That's because the aspect ratio of TVs used to be 6 by 4 and the end of court view fitted better. Now the aspect ratio is much wider, so you'd better a better picture if you put the camera in the middle, opposite the umpire. But, they've always done it like that...

We also need to examine our appetite for risk? Is it too low? Can cause innovation inertia.

Cloud doesn't change our business. Our business is not running IT. Our business is teaching, learning, research etc. Cloud is about changing the way we do our business, not the business itself. It's also about IT becoming more service orientated.
Much of the infrastructure is commodity now. If someone can do it better and cheaper why don't we let them.

We have to look at why we might want to keep things out of the cloud. Is it because we love the smell of a hot server in the morning...

the essential characteristics of cloud are:
On demand self service
Broad network access
Resource pooling
Rapid elasticity
Measured service

Tend not to get all of it, but bits here and there.

JANET 6 is our world class network which launches next week. It needs using. We should look at co-location. Get the tin out of the institution and somewhere else. (More than 50 miles away so people can't get in there cars and go and hug it.)

Several options from do nothing, to fully off site, via hybrid

UCAS uses cloud for burst capacity. Doesn't happen by magic. Took a lot of work to prepare UCAS applications to be cloud ready. But benefits huge.

Risk and innovation needs balancing.

What are the advantages of cloud?
Capacity, reliability, flexibility for large scale applications that are peaky
Cost effectiveness. Office 365 and Google are free.
Business Continuity
My comment, it's about service delivery and improvement
Cloud is about value, not cost

Barriers.
Mainly culture.
Some technical, ie getting applications cloud ready. But are lots of tools you can use.

Some competitive reasons eg HPC often kept in house. But why? Maybe culture. A lot of is commodity and people are buying it from amazon on their credit card as we speak.
JANET working on a provisioning portal for amazon

Some barriers are senior management who are concerned about cyber security




- Posted using BlogPress from my iPad

Sharing, collaborating and more awards.

In the HE sector we are very collaborative, much more so than in the private sector, despite being in competition for students and research grants. It's a theme that came up a lot in the last couple of days, as I've been out of Sheffield. On Tuesday I traveled to another University where I am a member of their new IT Committee. They have some major projects,  replacing nearly all of their enterprise systems, including finance, HR, payroll and student. Because they obviously like a challenge, they are using a best of breed approach rather than an integrated system to cover more than one set of functionality :-)  That's in addition to the other challenges we are all facing and other projects in areas such as learning and teaching and research. I'm there as an external to offer advice and guidance where I can, and give an outsiders perspective. Of course, it's not entirely altruistic, as there is a lot of learning to be done from visiting other institutions. In this case, I learned of the existence of the McKinsey Report into The Social Economy: Unlocking Value and Productivity through Social Technologies. I've had a quick look and the summary looks very interesting.

Yesterday I was at the Janet Brokerage Advisory Board. Well, that's how it started out, but as the funding which established the Brokerage was from the University Modernisation Fund which was time limited, we're now a Sector Advisory Board looking at Cloud Solutions. Extremely interesting and productive discussion, and one of the most enjoyable meetings I've been to for a while. We were looking at the way JANET, as part of the new JISC, could help Universities with adopting cloud solutions. What could be revenue generating (ie what would the sector be prepared to pay for), and what should be provided free as part of what you would expect from an NREN. We were looking at services, not just infrastructure -  SaaS, IaaS and PaaS (Software, Infrastructure and Platoform as a Service).  JANET are working on getting vendor agreements with our major cloud vendors (eg Microsoft, Google and Dropbox) to save HEIs legal fees and make services easier to adopt.  We also talked about HPC and how much of that could be in the cloud, and why we are building our own.   Collaboration with each other played a big part of our discussions - on procurement, on negotiation of contracts, on sharing facilities, services, good practice and business plans. All good stuff, and I look forward to seeing the business plan which is the next stage of the evolution of this service.

Last week I posted about Tim Birkhead's award for teaching together with the video that we helped with, and tonight it was good to go to a celebration in his department. They were also celebrating being awarded  Athena Swan Silver status which is great news for them. The Athena Swan charter recognises employment good practice for women in STEM subjects (Science, Technology, Engineering, Medicine). The University overall holds a bronze award, and this is the first silver award in the Institution, so well done on a good week of news for Animal and Plant Sciences.

Cloudbusting

On Friday it was the MMIT (Multimedia Information Technology Group, part of CILIP) conference here in Sheffield. I was very pleased to give the opening welcome, and as the theme of the conference was "Cloudbusting - Demystifying the cloud", to give a very short overview of our experience of cloud services. So, I did a quick run through of the challenges facing us -
-->

Consumerisation of IT.

• 24/7
• User interfaces
• Mobility
• BYOD
• User interfaces
• Social networking
• User expectations
• Critical nature of IT

No surprises there! My point was that all of these can be drivers to move to cloud based services, such as our move to Google. What's important is the quality of the service - it isn't about cost cutting. A recent benchmarking exercise shows that our email service is about 70% cheaper than a peer group, which is good obviously, but the real driver should be about service provision. A better service, available and support 24/7, more accessible, more mobile, and more innovative - google can innovate much faster than we can. Which is mainly a good thing, but not always :-)

I couldn't stay for much more of the conference, but the keynote speaker talking about Searching in the Cloud was fascinating.  How do we find all of the stuff that's out there, and a particular issue, how do we link internal and external information. It can also be difficult yo keep track of your own personal stuff when it's outside the boundaries of your local system.  How true!

The University of Google

Exciting session now, our own Andy Tattersall talking about our move to the cloud (Google) and how he's been using it to aid collaboration in an academic department and the shift from enterprise software to Google Apps.
Popular session, almost standing room only.

Why did we go to Google? Interesting to hear this from one of our users. Luckily, Andy's reasons are exactly the same as I would have given :-)
Because they do it better. Our students and staff demand more than we can give them. Data storage, better ease of access, better communication tools, more resilient systems, social platforms, better interfaces. Students were already using google apps.
Staff were using tools like Dropbox. We had small amounts of storage, and a perceived lack of social collaborative tools. Also propriety tools are expensive.
Moved in two phases, students first. More adaptable than staff.
Then staff, no opt out, all moved across.

Andy's journey began before the University's in 2007 by doing 3D sketch up of the library, and started a ScHARR library blog. Then YouTube Channel. In 2010 moved the enquiry desk form to Google Forms. By 2011 when everyone moved he had a lot of experience.

Encouraged department to start to build web sites, more staff can use google sites and blogs for creating small bits of content. A lot of integrated functionality in google apps, - can easily embed docs, calendars, videos etc. into sites.
University still at stage of early adopters, but in front of many. On Gartner hype cycle, we have people on all parts of it, some have reached plateau, some hurtling to trough of disillusionment. Need champions, (nice picture of Graham McElearney there).

Awareness raising sessions in SCHARR include 2 hour hands on workshops, 20 minute bite size sessions, screencasts, webinars and hangouts. Also university wide events such a Making the Most of Google Day.

Need to change old habits, such as moving away from client to Google interface, using Google hangouts for meetings, presenting sessions. Sharing ideas is happening, lots of collaboration going on.
Doing things differently. Will get different results, will make mistakes, but will get better. And as Google moves forward, so will the University.

Interesting in the Q and A, Andy got asked a lot more questions about the use of the apps, different sorts of apps, what happened if apps get withdrawn etc. When I talk about our move, I tend to get asked about the technology aspects, and that old chestnut, security and privacy. A different audience I suppose, and also a very different presentation.

Excellent talk, and it generated a lot of interest.










- Posted using BlogPress from my iPad

KITs, ICE and satisfaction.

This week has involved a lot of KITs. No, not little cats, but Keep in Touch meetings. Vitally important for building networks, finding out what's going on, and sharing ideas and experiences. I try and meet most of my senior colleagues for an hour once a month, and this week a lot fell together - Coffee Revolution has done very well out of me this week, but I feel slightly high on caffeine!

Monday was spent in a design team meeting for our New Engineering Building, and we all trooped outside to put different cladding samples out so we could see them in daylight - such a lot to choose from, but luckily we were all agreed, and even agreed with the architect's choice. I think the building will look stunning, and although this isn't a very good picture, gives an idea of what it will look like. Of course, the reason we're involved is the amount of student-led, IC type space it will have in it, including group study rooms and creative media pods which will complement what we have in the IC.

Yesterday we had a meeting to discuss the results of a recent staff survey with our senior managers. This was a University wide survey asking a number of different questions about staff satisfaction with their jobs, their department, their management and the University.  It's fair to say the results were mixed, some good, some where there's room for improvement. So, an action plan is being  drawn up, together with some facilitated discussion events so we can better understand the reasons behind some of the answers.

Today I've been taking to the Janet Brokerage Service about our move to Google, our experiences of contract negotiation, and some of the concerns other Universities seem to have. Hopefully they will be able to take the lead in discussions with Google, as they have done recently with Microsoft, to make it easier for more Universities to move to cloud based services.

Anything, Anytime, Anywhere

Here I am at the Gartner Portal, Content and Collaboration Summit. Opening welcome is slick as always. Covered the most important interconnected forces around today:
Information, Mobile, Cloud and Social.
Combine to create an "always on" era.

First keynote follows on from this and is entitled Always On. Anything, Anytime, Anywhere. I remember going to a SUN ERC conference about 12 years ago where I heard that phrase coined for the first time.

Industrialisation of IT does not mean the death of the IT professional, but frees us up to do more interesting things.

Picture of a typical CIO. Looking out over the horizon. Or, not sure what it is or how it got there.




CIO technology priorities at the moment include cloud and mobile, and are looking for customer experience to drive innovation. This latter one is a big change from previous surveys. IT more than ever now has to be part of the business, cannot be separate.

Social computing not very high on CIO list of priorities. (NB, it is on mine :-)). But, the only way we will deliver solutions to our business problems is by collaborating with each other.

How do we put the Nexus of the 4 forces, Information, Mobile, Cloud and Social, to work? Things we should be looking at to build really cool collaborative apps:
Image and position capture
Integration into the social network through APIs
Tagging
Approval workflow
Text and natural language analytics

Some other scenarios:
Second screen
enhance real works with virtual overlay. Eg watching a film with iPad on your knee to interact and look things up
Or meta data over a sports event.

Healthcare
Health management of the whole individual based on the individual's interactions inside and outside of a doctors office. People are more honest with their friends than their doctors. Big privacy and ethical issues of course.

People are driving a convergence of information, social interaction, mobility and cloud. Using multiple devices and applications of their choosing people connect with each other and interact with a wealth of information. Seamlessness of the experience relies on underlying cloud infrastructure.
People expect this interactivity in all of their roles, ie personal and work.
BYOD will be the norm. Followed by bring your own applications.
We used to talk about rogue developers. Now we have citizen developers.
We have to give up some control, in order to have some control.

People are drivers. Customers switch between open and closed environments, apps, devices etc, but all they see is the glass. They don't know, nor want to know what goes on behind it. User manuals barely exist now. Things just work. Or they should. Big challenge for our architectures which are often complex and based on legacy technologies.

What is a portal now? Integration of different apps? But iPads do that without a portal. Tying to solve problems in wrong place in the architecture. Too close to the glass.
Existing architectures are usually conceived in isolation and are techno centric not user centric.

Empowered individuals are untethered, work anywhere on anything. Have access to great design, technology and app choices, and will quickly discard apps that they don't like and move on. They aggregate and integrate information. They will engage with technology if it provides value to them.

Information is key to good apps. From multiple sources, historical and in real time, specific to the individual, relative to the individual's social connections, informed by their behavioural patterns. Interesting problems to work on.

Is our view of enterprise content management in our organisations informed by the above, or are we still document pushers?

Interaction across devices is important. We all use different devices for different experiences. Eg iPads for consuming content, laptops for creating. Therefore consistent state and data access is vital.

Lots of architecture implications for everything covered in this talk.
We need to embrace a post modern architecture,
Think in terms of an ecosystem. Mutual and interdependent. Co-creative, innovative and collaborative.
SOA is important.
APIs are the skeleton key, design APIs for your consumers
Choose from the range of app development tools
Legacy modernisation is the elephant in the room. It's required but not easy.
Architect for the Cloud.
Evaluate IaaS and PaaS providers
Understand the business. Build things that add value.

We need to channel our inner anthropologist, sociologist and ethnographer
It's not just about technology anymore!



- Posted using BlogPress from my iPad

Janet Brokerage advisory Group

Today I've been in Harwell at the Advisory Board for the Janet Brokerage. Lots of interesting discussions, can't report on all of them for commercial reasons, but will try and give a flavour of things covered.

The Brokerage are talking to major cloud service providers, and we had a report on negotiations with Microsoft about their 365 service. Work is being done on standardising contracts, terms and conditions, and different licensing offerings to the sector. Other cloud providers will also be approached, including Google.

We also confirmed what the eight suppliers on the framework agreement offer. In summary:
Collocation at tier 2 and above
IaaS, including VM ware, public, private, community and hybrid clouds
Combined procurement
Complementary services eg consultancy, planning and managed services.

Use cases include off site data centres, Disaster Recovery, hybrid virtual infrastructure, cloud burst and research infrastructure

There's an event coming up soon where we will be able to ask the suppliers what they can do for us, and for them to find out what it is we need. It's a speed dating event, and details are here.

The brokerage are also working with other bodies including UCISA on costing of IT Services, with the hope of producing some sort of calculator or modelling tool. Knowing the total cost of services is an important part of any cloud or outsourcing decision making process.

During the day I learnt about Moonshot, a project looking at authentication and federated access to applications in the HE community led by JANET. looks interesting, and of particular use by researchers.

Finally, we had a demonstration of meaning based computing from Autonomy. Using their Intelligent Data Operating Layer (IDOL), very intelligent and contextual searches can be carried out across multiple data sources, looking for patterns. Think searching emails from Barclays Bank looking for evidence of rate fixing.... The can also search media, including in socialedia looking for trends, looking for what people are saying about you, your brand, your product, and giving it a positive or negative rating. Unfortunately I had to leave to catch my train before the end, but was very impressed with what I saw, and could see an application for the intelligent and contextual searching in the research collaboration space.


- Posted using BlogPress from my iPad

Gardens, SSB, Portals and Cloud

Sorry for short hiatus in blogging - my annual visit to Chelsea Flower Show has interrupted this week. Great day, glorious sunshine and stunning gardens. If you're interested, my favourite was Chris Beardshaw's Furzey Garden, absolutely gorgeous. And for pure bonkedness, Diarmund Gavin's pyramid garden - a crazy tree house. Just a shame you had to be a friend of his to go up it!


Anyway, back to work related matters. Service Strategy Board last Friday had a good look at the recommendations coming out of our awayday about how we could make it work better. We'll be developing much better service descriptions, developing service level agreements with measureable things that we can use to monitor how well we're performing, and developing a strategy for each appropriate service area.  We're also going to carry out a revie of how work comes into the department, which currently is through many channels, including requests from Service Advisory groups, through our SER (service enhancement request ) system, throgh the helpdesk, and through ad hc meetings, conversations and telephone calls. the intention will be to streamline the process and funnel as many requests as possible through one route so thay can be properly prioritised and have resources allocated to them where necessary.

We also had a strategic liaison meeting with the Faculty of Science, where we had an interesting discussion about creating apprenticeships and training schemes for technical staff - something we're keen to work with them on.

I've also been to a catch up meeting on our new portal project where we've changed direction somewhat and are developing something in house which will achieve a lot of what we need to, and we will be looking to combine it with iGoogle to provide access to our services. More on that later, but it's looking very exciting.

Finally last week, I recorded a 10 minute presentation for a JISC webcast on cloud computing. I was asked to speak at it but couldn't make it, so we've combined a recording of me talking to a camera in my office with an Echo360 audio and slide recording, which I hope they'll be able to edit and show as part of the day. Always a bit weird being recorded, but was quite pleased that I took a 20 minute presentation and in one take got it down to 9 minutes 58 seconds!

This week so far has been mainly catching up, going to Chelsea, and  walking round our new building getting used to where everyone is now. I think it's great that so many us are together, and you can bump into so many people just walking to the kitchen or printer.

The Glitterati meet

I spent yesterday in Didcot at the Janet headquarters at the first meeting of the Janet Brokerage Advisory Group. On the way there I was amused to see a headline describing us as "glitterati" and "thought leaders". Not sure which one I prefer. Someone in the department was kind enough to point out to me that the definition of glitterati is "plural noun - wealthy or famous people who conspicuously or ostentatiously attend fashionable events". Very funny!

The brokerage has been established using funds from the University Modernisation Fund to assist Universities in adopting cloud technologies - I've been following progress with interest as I'm also on the HEFCE Steering Group overseeing this programme.

There are obviously many reasons for moving to Cloud (and we're not getting too hung up about our definition of cloud - keeping it fairly broad).  These include improving and modernising our services, improving efficiency, cost savings, moving staff effort away from providing commodity services and into areas which add value to the staff and student experience, becoming more agile, and keeping up with the pace of innovation.

The role of the Brokerage is to make is easier for Universities to adopt these technologies, to remove hurdles, promote good practice, assist with supplier negotiations and help with business cases, partuclary around costing of services. We discussed the latter at some length, especially TCO (total cost of ownership). How many of us know exactly how much our services cost in total. Do we ever include space, power, procurement, installation, commissioning. All of these are included when we look at the cost of outsourced services, so we need to be comparing like with like.

So far the Brokerage has established a framework with 8 suppliers for data centre and  cloud based services, but has a number of other priority areas including assisting with moving to cloud based collaboration tools (mainly Microsoft and Google), tiered storage, HPC in the cloud and TCO. We also had a discussion about they could help us with virtualised desktops, so that's something to look at.

The role of the Advisory Board is to provide independent governance, to give guidance and advice and to assist with information gathering from the sector. I'm looking forward to being part of it. When I presented to the department the other day on our key objectives for the coming year, I made it clear that we have to consider all forms of service delivery  in all decisions we take about our services, and that includes cloud. So, hopefully we'll one of the brokerage's customers at some point.

Cryer and Clouds

Conference dinner last night, so as usual, a slightly subdued start this morning! After dinner speaker was Barry Cryer, who was very good. Spoke for quite some time, told some great jokes, and managed to get a dig in at many of the people there, including me. Was lucky enough to have a chat with him afterwards, and find out what really happens when they're recording I'm Sorry I haven't a Clue!

First session this morning delivered by Logicalis and was about Cloud, and how it might help us transition to a customer centric experience for students. The technology is there, commercial pressures are there, and now we have the Janet Brokerage to help us procure these services.

Cloud is a mixture of technologies which we need to blend to get a solution that's best for us. Public cloud services are widely used anyway. Private cloud and Hosted cloud solutions are also now mature and need to be considered.

Most young people expect to be able to use technology wherever they are, on whatever device they have, which are increasingly becoming smaller. We need to provide an immersive learning experience. In research, we will have to provide easy access to research resources, again on mobile devices.

Good video here about where the future might be going.

Current way of delivering IT is not sustainable, too controlling. Our approach to the endpoint device that is used to access our services will have to change. We will have no control over it. I think we've already gone down that route. Connectivity is key. Has to be ubiquitous, but high quality and high density. No good having wireless coverage if it doesn't work. Network design is critical to deliver media rich services. Then we have to design services so they can be delivered to any device.

Real requirement for innovation in IT is in risk based security. We have to look at trust mechanisms, methods of authentication and assess risks. Our security gateways will have to change. Again, taking a risk based approach.

So, we need to give a rich experience to our users, to their own devices using risk based security and over ubiquitous high quality networks. To do this, we will have to utilise cloud much more than we are doing now.


- Posted using BlogPress from my iPad

Janet Brokerage announce framework suppliers

This next talk is on the JANET Brokerage Service. It's been up and running for about 6 months now, and its aim is to become a hub for good practice in the cloud arena for UK HE.  It is a unit within JANET working between HEIs and suppliers to add value.

There's an interesting white paper on moving email to the cloud on their web site.

They've also recently announced that eight suppliers will be working with them on a framework to deliver cloud and data centre services. Not all of them announced yet, but will be over next few weeks as they sign the contracts.  The ones they can announce today are Dell, HP, Eduserv and Liberata.

They hope to provide value to both HEIs and supplies by aggregating demand, negotiating prices and assisting with contract negotiation. They accept that one size won't fit all, and will be working with Universities to make things are used in the right way.

They are setting up an Advisory Board (which I have been invited to sit on), so expect a lot more information about this service.

Does the cloud pay for itself?

Next up is Oxford Brookes, talking about making the cloud pay for itself.

How much does the cloud cost?
Consumer view is that its free, eg iPlayer, mail. So, is the cloud free? No, not that simple! Even consumers have costs, connections, data tariffs, devices, etc. So, there is a consumer cost.
Also costs borne by organisations - we provide devices, connections, Janet, hardware, software, people, skills etc. Content is an immense cost.

So, the cost of the Cloud is a lot! But a simple RoI is not the right metric. Need to see it not as a cost, but an investment. Look for business benefits, driven by customer expectations.
Innovation is another key driver. Everyone wants self service, mobile delivery, anytime, anyplace etc.
Carbon reduction targets also important.
Operational excellence, adopt best practice - buy a solution and use it out of the box. In principle can do this with cloud solutions.

Oxford Brookes have gone with Google for staff and students. Are some cost savings, eg licence costs for exchange. Not a lot of savings for docs, hangouts, sites yet. There are other ways of doing most of these things, so they are providing business benefits but not saving money.

Other cloud examples at Oxford Brookes include Moodle VLE  and web content management system. Doing some experiments with virtual desktop on Google Chromebooks. Also have a mobile app, but again not saving money, because not turning anything off. Just providing different way into services.

So, there is a lot of cost efficiency, ease of engagement, innovation, anytime anywhere  etc. Not necessary saving money.

Might be a cost in terms of reputational damage if services go down. Will nearly always be us that take the flack, not Amazon!  Need an exit strategy.

Conclusion - cloud services are probably not saving us money, but are getting business benefits and innovation.

JANET Cloud conference, outsourcing your datacentre

Today I'm at the JANET Cloud conference in London, speaking later today on our move to cloud with particular reference to security.

First session is from Middlesex University. They have 52 staff, and 37 hot desks. Everyone hot desks, including the senior staff and they can book hot desks in half day slots. They also work from home, and come together for project work. It makes them realise what they have to deliver themselves and what they can get third party suppliers to do. They have outsourced their datacentre and server management.Their data centre is based with IBM, it is Tier 4 and they use 10 racks at 18kw. It's about 8miles from campus. IBM staff do the tape backups. Middlesex can get into it, but very seldom do.

Also have a DR centre with them in Warwick, only 2 racks, 80 miles away. They own some of the hot equipment, but rest is shared and has significantly reduced costs.  There is some small amount of equipment on campus to handle local stuff eg printing. Only 5 racks.

Have virtualised almost everything, about 90%. Use VMWare.

Data centres joined up by virgin network to push data around, not JANET. Main data centre is then joined to JANET.

Wanted to move staff away from looking after hardware, patching etc, and have them looking after applications. IBM monitor hardware 24/7/365. They do all of the patching, Middlesex look after applications. Remote support is provided out of India, patching done locally. They also patch the SAN. There's a service and change management process in place.

Service and contract management very important. Terms and conditions, change management, charging, DR plans, DPA, security controls etc all need agreeing and documenting.

Exit strategy is looking for tier 4 DC with low PUE, servers and storage on demand, delivered using multi tenanted architecture and connected to JANET. Because everything is documented, should be easy to move. Have a defined set of requirements. Will no longer own hardware. Very liberating!

Why have they done it?
Uncertain Estates strategy. Unsuitable space allocated, power and air conditioning not maintained.
Didn't want staff to be doing patching. Needed them to work on significant projects which deliver visible business benefits to students and staff.
Budget pressures.
These sort of technology skills can be obtained from outside the university, don't need to be internal.
Total cost of ownership much lower if outsource. It isn't cheaper to run DC in house if you look at all costs. Focus on institution cost, not just IT cost.
Global 24/7 aspirations of University not viable on current University staffing.

Issued tender in January 2009, contract in place September, migrations finished by July 2010, transition complete January 2011.

Lovely quote from Paula, " moving a a data centre's a bugger".

Lessons learnt:
Shape and know your service contract- don't rely on supplier to do due diligence
Know your baseline configuration and technology, it's your exit strategy
ITIL practice is key

The cloud is maturing and stoppable, plan for it.
Why put high bandwidth services on campus, eg video content, if your students and staff are accessing them from off campus

For users it's the service that's important, not the location.

Need to help technical people let go, show them that there 's more interesting roles to have. Letting go is hard, and trusting a third party to do stuff you've been used to doing is hard. But it gets easier with practice!

- Posted using BlogPress from my iPad

Community clouds

Good meeting this morning being brought up to date on YHMAN's virtual shared data centre project.  We're running a pathfinder or pilot service, with machines based in three Universities, hosting virtual services based on VMware to form a community cloud. It's connected by a resilient ring with virtual security in place and is serving the eight universities in the MAN. You can see some slides about it on the YHMAN projects page here.  The technology behind it is all very clever - resilient and secure, and as we move forward with it we'll be looking at the use cases for it. Offsite back up, resilient storage, business continuity and temporary capacity are all possibilities for us. We're one of the current sites where a server is hosted, so we are both a supplier and customer. Quite exciting - will be interesting to see how it develops.

This afternoon we had a Business Continuity Operations group, looking mainly at our work programme for this year, and lessons learned from recent incidents and near misses.

Also had a phone call with a Gartner analyst, an expert in business process management, about some aspects of our implementation of service management. I find these conversations really useful, a they often offer a different perspective, especially as the Gartner analysts have a lot of experience and examples of good practice from all sectors - not just HE. They also know the questions to ask to make you think differently.

Working Differnetly Conference continued

The remaining comments from last week’s conference on new ways of working.

I went to a session on Cloud, given by VMware.  Not surprisingly, the general theme was that Server Virtualisation is the first step on the journey to cloud computing. Some other points:

• Cloud isn’t a new technology, it's a different approach to delivery. It enables the shift from in-house capital intensive IT, to the consumption of utility-based computing resources on an as-needed basis with an appropriate pay as you go model.
• Cloud mimics the delivery of a utility eg electricity. For example, if you had your generator you’d have to deal with maintenance, capital expenditure, consumables, and wondering with whether it could cope with the surge when you turned the Christmas lights on.
• Cloud has 5 characteristics:
• On demand, network access, resource pooling, elasticity, pay as you go.
• There are 130+ data centres in central government containing 90,000 servers running at 7% utilization.
• A combination of virtualisation and cloud should drive down IT costs which you can reinvest in other parts of your business,  and it should also increase agility.
• Automation v important. Amazon have 1 engineer for 1500 servers.
• The New world is using mobile devices. Users are demanding access to apps and data on the move. Desktops are expensive to refresh. Solve this problem by unlocking data from the desktop and putting it in the cloud whilst keeping it secure.

The next session I went to was on Security in a Mobile World, given by Sophos.
Some notes I took at the time:

• We've lost the argument about connecting things to the network. People are bringing their own devices. Applications are increasingly being delivered by a browser.
• Smartphones and tablets are scaleable and increase productivity. They are cloud ready, and they are cheap compared to a corporate laptop.
• Do they cause problems? Operating systems change frequently. Windows tends to be stable, but these update themselves.
• Then there’s app stores. How do you stop people downloading stuff like Angry Birds to corporate devices (why would you want to?)
• Most organizations Acceptable Use Policies are inadequate and poorly used. Most were not written with mobile devices in mind.
• Executive teams are prepared to accept the rise of these devices. There are 70m Blackberries in world and 90m iPads.  There’s been a massive growth of not-enterprise ready devices.  Will the use of them get regulated, eg by ICO?  They will unless we take this seriously and deal with it.  Hmm, I don’t agree with this. You can tell this session is being given by a security vendor.
• Mobility and consumerisation will only work if it's secure. Secure working practices require a whole company approach. Can't just be legal dept or IT dept.
• Consider the implications of the power of the device you've got. It's a tool to do your job. Treat them as tools of the job. You must demand individuals are accountable for their actions. Consideration must be given to what happens when things go wrong. In the main, failure is almost human not technology.
• He’s skeptical of cloud because of security.

So, one session on how cloud and mobile is going to save us, and one telling us it all needs regulating. Well I suppose that’s balanced