Nothing New

Apologies for not actually having posted anything in a while. Not feeling too hot at the moment but am drafting new posts which will be uploaded Monday.

Frustration of lack of understanding from Management

It becomes frustrating when I look at on-going work I am doing and that the order of things does not appear to be the right way to do this, but due to the way the business works the order has been dictated and now trying to change this will be an impossible task.
Currently I am carrying out the following in this order:
• Archiving Implementation (EMC Centera)
• VTL Implementation (Data Domain)
• VMware Backup Solution (Veeam)
• SAN Fabric Merge (Brocade)
• Storage/Backup Review
• New Data Centre (Nexus)
• Data Protector Upgrade
• CommVault Upgrade
• Storage Management Software (DPA/IONIX)
• Storage Consolidation (VMAX?)

As you can see there is a lot going on coupled with a number of Oracle Application Projects I am involved in this amounts to a huge amount of work and really shouldn’t be rushed. So what is the correct order well I have suggested a roadmap with the tasks and also more minor ones set out over the next few years and it doesn’t really bare much resemblance to the current order.

The problem is I work for a Retailer where IT is a hindrance more than an enabler to the business so when they see something they want it there and then and it doesn’t matter so as long as it doesn’t impact service.
The order that really should be followed which I have shown and detailed but the powers that be seem to agree then ignore the recommendation is as follows:

• Storage/Backup Review
o At least know your starting point and what needs fixing.
• SAN Fabric Merge
o Having 2 separate SAN’s is a blocker to most of the items
• Consolidation of Backup Software
o You will notice this is a new item, but I don’t see the need to have 2 separate Data Centre Backup products in Data Protector & CommVault. So let’s standardise on one, probably not either of the incumbents but something proven like NetBackup or even TSM.
• Archiving Implementation
o Well this utilises the backup software so can only follow a piece of work to the software which is the enabler for archiving otherwise you won’t see the correct results and I have tried to tell people this till I am blue in the face but will they understand.
• VMware Backup Solution
o Once this is deployed then the amount of data needed to go to tape or VTL also reduces.
• VTL Implementation
o Once Archiving is in full swing then there is less data to backup which in turn lessens the load on the VTL and allows for shorter backup windows and also helps with recovery.
• New Data Centre
o Current restrictions on anything being placed in the Data Centre so can’t address new storage arrays until we have more space.
• Storage Consolidation/Storage Management Software
o I see these 2 as closely linked

Only time will tell if the current order will cause problems but at the moment problems are arising all the time but all the guys at the coal face get told is to make sure things keep running, so it’s a fine juggling act between BAU and Project work.

The big question I have is why bother having an Architect looking at this for you if all you do is ignore them and plough ahead anyway without any real outlook?

SAN versus Nexus

This issue keeps appearing in fact on a weekly basis due to the fact 2 projects I am involved in touch this issue from different angles.
A piece of context the 2 Projects are:

• SAN Merge – Currently Lead Architect on merging the 2 disparate SAN Infrastructures the current company I work for has to deal with on a daily basis, the problems of which are too many to go into in this discussion. But basically this is a merge of 2 sets of Brocade Infrastructure and replace with DCX Switches to give performance and capacity increases.

• New Data Centre – Currently Lead Architect on building a new Data Centre due to the fact that the current main Data Centre is full and has no room for growth in respect to power, cooling or space. This is looking highly likely will go Cisco Nexus to provide the new Network Infrastructure.

Now you may be thinking well what’s the issue, well the problem arises when you look at detail into Cisco Nexus and the whole unified fabric options from Cisco which of course in turn reduce, cabling, power, heat and space. But of course if you explore this option then why would you have a separate SAN Fabric for Fibre traffic and spend £m’s on this when the Nexus platform is the enabler to not have a separate SAN Infrastructure. Of course the Arrays currently held within the existing Data Centres do not have the ability to be FCoE, but then when I think ahead 6-12 months there is a piece of work in it’s infancy to look at consolidating the Array Infrastructure and in turn replacing it with newer best of breed technology and I am think VMAX here potentially with another providers storage acting as a 2nd maybe even 3rd Tier.

So what is the best way to handle this at the moment 2 steam trains (which I use as an analogy for the 2 projects, due to the fact the PM’s are pushing ahead without thinking of the bigger picture) are heading down 2 separate tracks in the same direction.
I can see it getting to the stage at the end of the year where we have spent £m’s on a new Brocade SAN Fabric and £m’s on Nexus for the new Data Centre and in January when we start to look at putting hardware in people then realise well hold on we don’t need to utilise those Brocades as we can hook straight into Nexus to provide the uplinks to the Blades. Then that’s when the question will come “Why the hell did we do both?” and I will be tempted to sit in my “Ivory Architecture Tower” and reply “I told you so”. But will I really say that, or will I take the hit and think of something wonderful to utilise both spends and justify why we did both; of course if we keep the old Arrays then I have a reason for doing both as I still need a separate SAN Fabric.
So maybe that’s the solution to the whole problem, keep the company in the dark ages in respect to Arrays to make sure no one notices we spent double the amount we needed to by implementing both solutions.
There is the wider argument that Nexus would be a mistake to fully utilise and it is always best to keep Fibre Traffic Segregated on SAN Switches, but I am not so sure that is the case anymore.

Are companies ready for full unified Fabric’s, well some are going that way and Cisco seem to be making a good crack at convincing people that it’s the way forward. But to be honest if the Business Case stacks up and you can’t justify the need for Separate FC Fabric then why not.

First Attempt

Now I must admit, I have been following a number of bloggers for some time and always said to myself “never will I find the time to do that, and what would I even talk about or discuss”. But then I realised working in the cynical world of IT as I do, I probably have a great deal of information holed up in my head waiting to come out. The bigger question is does this just contain a melee of useless information and opinions that no one really wants to hear!

Where do you start with your first blog do you launch head strong into a detailed opinion or comment on something out there in the market that you either agree or disagree with?

Well I decided that it may not be the best idea to do that from the word “Go”, so perhaps a more detailed account of where I have come from and me as a person will help set any future blogs into some sort of context. I have watched many bloggers over the years be ridiculed for actually working for the vendor they always seem to promote. Then the bigger question that arises is “can you ever be truly objective about anything if your livelihood depends on it?”

I will start with the here and now, I am currently a Technical Architect for a Retail Supermarket chain in the UK called Wm Morrison plc, been here 18 months and starting to assess my position and where I want my career to be long term. Previous experience has heavily been within financial institutions and also the Oil & Gas industry. Architecture wise I am Togaf 9.0 certified and the bulk of my skills and knowledge is around the Data Centre and Infrastructure, but with a smattering of Oracle applications and Web technology.

I like to understand technology and the benefits it can bring to organisations and always look at the bigger picture i.e. can what I buy today help me in other areas besides the primary reason I buy it.

What frustrates me is Vendors who cover up aspects they don’t want people to find out and show a nice glossy piece of technology and try to ram it down your throat even if it’s not even right for you. Also white papers that clearly have a product slant on them even though they claim to be “non-biased”. I suppose that could be the reason why I am starting a blog to give my opinion which is only my opinion take it as you will.

It has been said I work for EMC by a number of people who have seen my tweets or who have seen me at technology events, but I don’t it just so happens I have been heavily involved in EMC technology for some time and I see its merits, but at the same time I understand the flaws in some of the offerings or even in the way the company tries to sell them.

Well I think that’s not half bad for a starter for 10, only time will tell if I am wasting my time, I suppose the real crunch will come when I post my first blog about a piece of technology or opinion rather than a quick run through of me!

EMC & Greenplum

Well the EMC behemoth trundles along like an unstoppable train and picks up another company in the shape of Greenplum. That opening statement is open to interpretation as to be honest do I really think EMC is an “unstoppable train”; no not really it is a clear and calculating agile company that only picks up pieces along its journey that ultimately form an end goal. Currently I think the end goal is to be the Cloud partner of choice but then I could be wrong.

So Greenplum well were to start, with regards to DW DB (Data Warehouse Database) Managements Systems its getting there if the Gartner Magic Quadrant from January 2010 is to be believed and to be hones they are not normally too far off the pace. It’s an MPP data warehouse DBMS based on open-source and has just over 100 customers worldwide currently (you would expect that to increase now it comes under the EMC umbrella).

It was the first company to deliver a DBMS solution for use in a private cloud infrastructure and people like Teradata have since followed suit. It regular competes against IBM, Oracle and Teradata and probably has lost out in customers before due to the fact it has a small customer base and hasn’t had a huge push behind it but guess what it has now.

So how do I see it fitting into EMC, well as I don’t work for EMC my thoughts are purely speculative, so a read of Chuck’s blog will maybe give some insight as I see he has already posted something on this topic.
http://chucksblog.emc.com/chucks_blog/2010/07/emc-to-acquire-greenplum.html

The plan as I see it from reading press releases and looking at Greenplum’s own website is that it will become the foundation of the new Data Computing Product Division.

I actually see Greenplum being absorbed into the EMC family very well which can only benefit Greenplum and its existing customer base. The amount of available R&D funding will increase significantly which will allow for all those fixes and enhancements that Greenplum pushed to the back burner to be brought forward. The likes of Netezza will now have a competitor with huge backing and could see their market share start to be impacted, as for others such as Oracle and IBM I am not sure they will see any major hit against them in the short term. But then again who knows it all depends on how the offering is marketed and enhanced by the overall EMC product set.

The big question that springs to my mind and maybe I am jumping the gun slightly but who will be next, I have a feeling maybe incorrectly that there will a bigger purchase by EMC before the end of the year to take their overall cloud offering even further forward.

So watch this space as they say………