Jul 222016

Hello readers, its been a while since I’ve done a blog post and I have to acknowledge I’ve been rubbish keeping up with my blogging duties.

However this is still a good medium for me to let you all know what happening with your favourite Mackem.

After 4 GREAT years with VCE (and a year and a half reselling Vblocks before that) I’ve taken the difficult decision to move on. I’ve seen VCE grow from a $100mil a year in revenue to over $2 billion. No matter what you think of VCE, there can be no doubt those figures are staggering. VCE created and lead the CI marketplace from day one and has driven huge change in how infrastructure is consumed. The values of simplicity, agility, lower TCO etc are the cornerstone of VCEs CI offerings way before Cloud was being adopted in any mainstream way.

(one of my babies)

It should almost go without saying that while the company had a great strategy, it was the people that made it happen. I can hand on heart say that I leave behind some of the smartest and most driven people I’ve had the pleasure of working with in my career to date.

Under the stewardship of VCE, in the 4 years I worked there I was promoted twice and was a member of many internal initiatives to help drive things like integration into the wider EMC machine and optimising current practises. Also with my side role as a Technical Team Lead I got to mentor some inspirational youngsters (and some who wish they were young) who will be forces to be reckoned with in years to come.

So WHY leave if it was so amazing, I hear you all cry.

I’m sure you will all agree, IT has evolved faster and more dramatically than ever before in the last 4-5 years. Trying to remain relevant as an individual is becoming harder and harder in a world where infrastructure alone is becoming less and less relevant as a differentiation for customers. As a fiercely ambitions person I feel the role I was offered (its a nice feeling to be headhunted lol) gives me an opportunity to take a career step up and also open doors for serious progression over the next 5-10 years.

 Oh yes, the world will be mine (evil laugh)


So to that end, I am happy to announce that I have joined Oracle Cloud as a Cloud Leader specialising in IaaS.

My new responsibilities from 30,000 feet include Evangelising, helping defining strategy and Go to Market, discussing and building business and relationships associated with the Oracle Cloud & Customer programmes, the Oracle Cloud Machine and Oracle’s highly competitive IaaS offerings. I’m sure some of you vExperts reading this will be aware of Ravello, ill be focusing on that specifically and I plan to do lots of posts on that super coolness in the coming weeks and months. What’s cool is this role is flexible so ill be able to go as high level or as deep as I need to.

There are a HUGE bunch of things I’m super stoked about by joining Oracle, not just is Oracle a Cloud Provider but its one that covers all the bases in a way others currently aren’t. Its also a massive challenge to try and take on the market leaders, but its one that I relish. Oracle are SERIOUS about cloud, they are investing and WILL be one of the major players in the public and hybrid cloud space including of all flavours cloud in the years to come.

THE STANDING JOKE: Those who already know I’m now part of the ‘big red machine’ have already ribbed me about licensing. I don’t think that can be avoided so feel free to send me the Jibes Smile

Fingers crossed my friends and followers won’t now abandon me and hopefully you will find my insights in coming months and years useful and interesting.

Post to Twitter

Oct 222014

One thing was obvious at VMworld Europe this year, the amount of solutions based on end to end pre-defined architectures was staggering. Converged Infrastructure (CI) has never been so high on the agenda but with the announcement of EVO:RAIL CI now on the lips of everyone from C-Level execs to the Joe Admin at the coal face.

The Official Dilbert Website featuring Scott Adams Dilbert strips, animations and more

This was a real eye opener for me personally. I have spent the last 4 to 5 years on booths trying to explain to people what CI is and why its not just a ‘marking’ play. (though some CI is just that). To have most people come to the VCE booth this year already with the concepts behind CI in their heads and how it can benefit their business was fantastic and allowed for real solution focused conversations. This was the moment that really showed me that CI here to stay and is how companies will deploy their infrastructure in the future.

Pretty much everyone that came to the booth had a similar message from walking the floor though. “Every vendor is saying they have THE ANSWER, but not just one answer they have several, im so confused”. After walking the floor and knowing the CI and wider industry how I do, i cant say i blame them. There is so much choice that people are stuck seeing trees but missing the wood.

I came across this quote on wikipedia from The Paradox of Choice

Autonomy and Freedom of choice are critical to our well being, and choice is critical to freedom and autonomy. Nonetheless, though modern Americans have more choice than any group of people ever has before, and thus, presumably, more freedom and autonomy, we don’t seem to be benefiting from it psychologically.

—quoted from Ch.5, The Paradox of Choice, 2004

This really does ring true with the current state of the industry IMO. CI is bad enough right now as a competitive space with different ways to cut the cake, but add in all the IaaS, PaaS, SDN, SNS, 3rd platform this and that etc that everyone and their dog are offing a layer on the cake and you have a critical mass of confusion. Even vendors that offer a single 2u CI offering are bundling in everything under the sun as standard and then some more on top as options.

How companies approach support, operate, maintenance, upgrades, patchs etc etc when faced with such complexity is without doubt the challenge. Speeds and feeds along with general capability are obviously important but even if you treat the tin as commodity those soft issues are not to be shirked and are the main concern at large enterprises and SPs, especially at the C-Level.

One of the benefits of going CI is simplicity both from a tin/software perspective but also from a high level ops view. I’m always honest with people I talk to. If, for example, you want the latest Haswell Chip the day after a Cisco GAs it then VCE and Vblock may not be for you as it undermines the values of Vblock and increases the risk. Not every version of every product from the VCE investor companies (yes Cisco is still and investor before anyone things of a cheeky comment) can be integrated as we would be testing for in an infinite loop and never release a product. So we streamline those choices while trying not to impede on that autonomy and freedom that choice brings.

The market in general is in a state of flux. No one quite knows what the future will hold. Its becoming quite obvious (to me at least) that the whole world will not be dominated by the likes of AWS and Azure anytime soon (perhaps when my son starts work it will be, he’s 6 btw). So there will undoubtedly be some natural thinning in the marketplace i feel in the near to mid future. Bets are being hedged and people are starting to make decisions but in my humble opinion companies will be struggling with Confusion, Complexity and vendor Sprawl for some time to come.

Obviously, i think im work for a company with the right approach. But this is not a sales post, just a general observation of the market and the complexity companies are facing on a day to day basis.

Post to Twitter

Dec 232013

I know the title is a little cheezy… sorry 😉

An of advantage of Vblock using Panduit as its cabinet supplier means the custom cabinet and cable management components that Panduit supply to VCE allows for optimal space and cable management based around a set exacting standards. It allows for maximum efficiency when building out the physical components into the cabinets.

However, there are sometimes DC requirements that go beyond what is a standard option within the Vblock configuration. Being that Panduit don’t just supply cabinets but a whole range of DC options means that VCE can meet almost any DC requirement no matter how demanding.

One of my customers who use a Colo DC required that the Vblock must be supplied and fit into a cold aisle containment row. The DC did have a supplier for their default cabinet cold aisle containment but unfortunately that company had gone out of business.

We deployed the Panduit Net-Contain Containment System.


Panduit and VCE worked together to scope and deploy the Vblock and Cold Aisle Containment system.


Those familiar with VMAX with know that its mandatory to use EMC cabinets are are actually a different height from the Panduit cabinets used in the rest of the Vblock. Panduit produced custom built panels at the top of the VMAX Titan racks to fill the void and ensure non of the cold air escaped. Floor panels designated for expansion are also covered with floor to roof blanking panels which are easily removable.

Post to Twitter

Jun 212012

When following the Cisco Cheat sheet for deploying the Nexus 1000v for deploying the Nexus 1000v in L3 mode, make sure one of the first commands you run (or when applying your config) is ‘feature LACP’ without it you cannot activate your LACP port channel.


I set up something similar to the above, the port-channel is between the server ports and the 5k (2ks acting as fabric extenders) and setup to load balance using LACP. This is design was only recently supported by Cisco as previously you would have had to load balance traffic on the port channel via MAC-Pinning.

This annoying little oversight caused me an hour of grief on a customer site as the port-channel on the 5k and 1k would not become active until this feature was enabled.

To enable this feature:

n1000v# config t

n1000v(config)# feature lacp


Post to Twitter

Feb 132012

I recently came across a customer that had limited space, power and cooling in there datacentre (very few who don’t right?) but wanted to put in a Vblock but to do so would need to split most of the UCS chassis across multiple racks and at opposing ends of the datahall. Traditionally, when I design and spec a UCS system I use the default ‘passive‘ Copper Twinax SFP+. In the event I need to provide cabling for Fabric Interconnects that are more that 5 meters away from the chassis then I would use ‘active‘ Copper Twinax SFP+ as these can go up to 10 meters.

But in this case the distances are over 30 meters. The alternative is go optical by using SFP+ modules (SFP-H10GB-SR) which can more than compensate for almost any datacentre distances (300m or so).

A few of you may have noticed I said this was a Vblock, you may be thinking you will not be aloud to do this with a Vblock as it brakes the default spec. While it does go against what’s in the design for a Vblock, it is a great example of how the Vblock products are actually flexible and not as rigid a people may think and exceptions can be raised when genuine requirements demand it.

Click here for more info on installing and configuring a UCS chassis and cabling it up.


Thanks to a Andrew Sharrock (@AndrewSharrock) for pointing this one out. As of UCS Software release 1.4 Fabric Extender Transceivers have been supported and are an alternative to using the above. You can get up to 100m from a FET and it supports OM2, 3 and 4 cables. I have a feeling not many people have deployed this as its Google doesn’t bring many results back on this subject but its an option. I’m not sure if VCE support it within a Vblock either (VCE peeps are welcome to confirm or deny this in the comments).

From Cisco:

To replace a copper Twinax SFP+ transceiver with an optical SFP+ transceiver, follow these steps:

Step 1 Remove the copper Twinax SFP+ from the I/O module port by pulling gently on the rubber loop (see Figure 2-19). The cable and SFP+ transceiver come out as a single unit, leaving the I/O module port empty.

Step 2 Insert the optical SFP+ transceiver into the I/O module port. Make sure that it clicks firmly into place.

Step 3 Plug the fiber optic cable into the optical SFP+ transceiver (see Figure 2-20).

Figure 2-19 Removing a Twinax Copper SFP+ Transceiver

Figure 2-20 Replacing a Copper SFP+ Transceiver With an Optical SFP+ Transceiver

Post to Twitter

Nov 212011


Click here to download.

Here’s a recap of the new hardware and software features that was introduced in UCS Manager 2.0.

New Hardware Features in Release 2.0(1)

Release 2.0(1s) adds support for:
– Intel Xeon x5687 CPU on B200 M2
Release 2.0(1m) adds support for:
– Cisco UCS 6248 Fabric interconnect
– Cisco 2208 IO Module
– 2500 Watt DC Power Supply for the Cisco UCS 5108 Blade Server Chassis
New Software Features in Release 2.0(1)

Release 2.0(1) adds support for:

-Licensing – Updated information for new UCS hardware.
– Firmware Bundle Option – Enables you to select a bundle instead of a version when updating firmware using the Cisco UCS Manager GUI.
– Disk Drive Monitoring Support – Support for disk drive monitoring on certain blade servers and a specific LSI storage controller firmware level.
– iSCSI Boot – iSCSI boot enables a server to boot its operating system from an iSCSI target machine located remotely over a network.
– Pre-login Banner – Displays user-defined banner text prior to login when a user logs into Cisco UCS Manager using the GUI or CLI.
– Unified Ports – Unified ports are ports on the 6200 series fabric interconnect that can be configured to carry either Ethernet or Fibre Channel traffic.
– Upstream Disjoint Layer-2 Networks – Enables you to configure Cisco UCS to communicate with upstream disjoint layer-2 networks.
– Virtual Interfaces – The number of vNICs and vHBAs configurable for a service profile is determined by adapter capability and the amount of virtual interface (VIF) namespace available on the adapter.
– VM-FEX Integration for VMware – Cisco Virtual Machine Fabric Extender (VM-FEX) for VMware provides management integration and network communication between Cisco UCS Manager and VMware vCenter. In previous releases, this functionality was known as VN-Link in Hardware.
– VM-FEX Integration for KVM (Red Hat Linux) – Cisco Virtual Machine Fabric Extender (VM-FEX) for VMware provides external switching for virtual machines running on a KVM Linux-based hypervisor in a Cisco UCS instance.

Post to Twitter

Nov 182011

A big thanks to my colleague Stuart Street for helping put this one together.


The Scenario:

Existing VMWARE 4.1 Cluster. 5 blades

Environment 1.3.(1t)

BIOS S5500.1.4.1f.0.120820101100

We needed to add a further 3 blades to the existing cluster.

So, the plan was to copy (clone)  from an existing service profile, create a firmware and boot policy on the new blades.  Or from the service profile template, create a new Unique service profile.

Next we diligently checked the various pools for enough spare entries. (IP Mgmt – UUID – MAC address – WWPN – WWNN). This was confirmed.

Trying to create the Service profile failed. Initially we saw cryptic error messages, leading us to believe it was mgmt. ip  or vnic  pool related.  The second error message does refer to a pool based issue.

SP creation method:-Cloning Service Profile

Error cloning

lsServer: can’t contain: object of class vnicIpV4PooledAddr with RN ipv4-pooled-addr, DN os org-root/ls-SERVERNAME/ipv4-pooled-addr.

SP creation method:-Creating Service Profile from Template

ucs error creating service profiles from template. Cause: pooled address is unknown

There were no clues elsewhere, as the service profile creation just fails and therefore the profile takes nothing from any of the pools.

So, we manually rolled back the firmware on the new blades before trying to create a new SP. But we could not do the BIOS however, unless there is a spare existing SP you can use to create a BIOS policy. In our case there was, so tried this BIOS version downgrade, this did not fix the problem.


The exact fix – Create a new UUID pool, add enough UUID’s – Point the service Profile at this new pool – result – clone or template method now both work.

“Explanation” is that somehow, rolling the firmware back on the Blades results in the existing UUID pool entries normally eligible becoming stranded for some reason.  Below is the link to the Cisco TAC explanation.  This  is not quite really the scenario we saw at all, and the labs at Cisco have not reproduced it  – but the fix worked!  We did not delete the existing UUID pool, didn’t see the point in taking any risk. Also the Fabric Interconnects were at 1.3.(1T) throughout.


(NOTE: you will need a Cisco ID to see the above)


So if you get any trouble with Service profile creation, on my list would be to start with creating a new UUID pool, referencing that new pool in the SP.  I probably would do the same with the other pools one by one as well in case this leads to resolution.. can always delete these if it turns out  they don’t help.

Post to Twitter

Nov 152011

PHD Virtual Backup Replications new version of the product seems to have some cool features.

I have picked out a few highlights:


8 Concurrent Data Streams per VBA – Twice the number of data streams allows each VBA™ to process up to 8 backup and restore processes concurrently for maximum performance and throughput. (Enterprise Version)


Optimized I/O Processing Engine – Enhanced I/O engine optimizes job processing providing more linear performance improvements when using multiple data streams and multiple vCPU resources.


Replicate Without Impacting Production – VM’s are replicated from your existing backups and backup storage without the need to perform additional snapshots. Other products require separate snapshots for both replication and backup, consuming additional resources and impacting your production environment.

Intelligent Data Replication – Intelligent Data Replication leverages deduplication and compression to minimize the amount of data transmitted to the secondary site. During on-going replication, only the data that has changed on production VM’s is replicated,
optimizing performance and bandwidth utilization of data transfer.

Replication Seeding – Initial replication of VM’s may be “seeded” to the DR site using portable drives to minimize data transmitted over the WAN. After initial “seeding” only changed data is replicated over the WAN.

Verify Replicated VM’s – A built in Test Mode enables you to validate replicated VM’s in the standby environment so you can be confident in your ability to successfully fail over in the case of a DR event.


OpenExport – Export VM’s from your existing backups for long term or offsite storage. VM’s are exported in standard compressed OVF format for portability.

OpenRestore – Restore your virtual machines directly to your VMware hosts without the need to install PHD Virtual software.

Mass Restore – Create and configure a single job to restore multiple virtual machines at the same time to reduce restoration time and complexity.

Optimized Tape Friendly Backup – You can export standard, compressed virtual machine OVF files to provide efficient and optimized long term storage to tape. Virtual machines can be directly recovered to a host from your tape backups.

Post to Twitter

Sep 152011

Using Hyper-V’s extensible switch framework the Nexus 1000v, as when Windows 8 is released, can now be used to create advanced virtual networking features within a Microsoft Hyper-V environment. VM-FEX (Hypervisor Bypass) will also available, its worth noting you have to be running Hyper-V on Cisco UCS utilising the VIC converged network card to get VM-FEX.

Check out the official announcement:


Its fair to say allot of people simply haven’t considered Hyper-V due to the lack of Nk1V support so it will be interesting if this any impact on VMware’s dominance.

Post to Twitter

Twitter links powered by Tweet This v1.8.3, a WordPress plugin for Twitter.