Sign up for my weekly newsletter, its short and sweet.

Category : Uncategorized

vxrack node

VxRack Nodes

I’ve been asked quite a bit lately on VxRack node types and options so I wanted to go over what is available and some of the options. First lets start off with the VxRack 1000 Flex options. At a high level there are 3 vxrack node types that can be configured in a VxRack solution; Dense Compute, Dense Storage, and a Storage only node.  Each of these nodes have a couple variations in order to properly size and accommodate various workloads.

VxRack Node Types 1000 FlexVxRack Node: Dense Compute

vxrack node dense compute

VxRack Node: Dense Storage

vxrack dense storage node


Winning Powerball with Big Data

If you are looking for the magic formula to win the lottery you have come to the wrong place. However today we will have some fun with some of the Powerball’s previous winning numbers so we can better decide on how we want to make our picks for the upcoming drawing.

Before we get started lets take a step back and get a basic overview/understanding of how powerball works. Drawings are held Wednesday and Saturday evenings at 10:59 p.m. Eastern Time. The game uses a 5/69 (white balls) + 1/26 (Powerballs) matrix from which winning numbers are chosen. Each play costs $2, or $3 with the Power Play option.

I also want to note that there was a format change on October 4, 2015 which we will want to keep in mind when calculating powerball winning historics. The white-ball pool increased from 59 to 69 while the Powerball pool decreased from 35 to 26. The new setup was designed to help create more winners and also more rollovers. Which is why we are now looking at a record estimated $1.3 billion dollar prize.

So the more and more I’ve researched different ways to predict powerball numbers using big data I was able to come up with two important facts that will save you some time and heart ache.

  1. The Powerball numbers are drawn randomly
  2. As humans we pick numbers non-randomly

About 70% to 80% of powerball purchases are computer picks and about 70% to 80% of winners are from computer picks. [source] So if I was going to play I would buy mostly randomly generated tickets based off of a computer; but possibly play one based off of historical data. I built this excel spreadsheet [download] with the data from on all historical winning numbers back to 1997. So download and enjoy (if you win I would expect a small donation 😉 )

In the excel spreadsheet you notice I have setup 3 tabs. The first tab is a random number generator for each of the white balls and the powerball in the event you don’t trust the powerball computer picks.

generate random powerball numbers

The second tab is a table of all of the past winning numbers in the order they were picked. This would be a good data source if you wanted to utilize some data processing to calculate odds and trends.

The third tab is the frequency of each winning number called and broken down by white balls and power balls. Here are some screenshots:


As I started crunching the numbers and odds I realized that the overall odds of random are already so high that if I want to actually play the powerball my best bet is to just let the computer pick the numbers. The moral of this unfulfilled blog entry is that we pick numbers non randomly and the powerball is picked at random.  So if you want to win, make sure you do it randomly 😉

Let me know your thoughts on the excel data and how else we could make the numbers more useful in the comments section below.

vmware vcloud air

VMware vCloud Air from Beta to Now

VMware OnDemand Early Access

I was given the privilege to test out what was formally called VMware vCHS on demand, and up until the beta phase of OnDemand you would have to size and buy chunks of resources if you were going to use the VMware vCHS service. When they announced the beta on demand program I was nothing but excited and intrigued as to how this new service would pan out.

Having the ability to spin up cloud resources ondemand from within a familiar interface (VMware vCenter) is everything a virtualization admin could ever want. I had my doubts about how easy it would actually be though. I assumed much coding or special plugins would be needed in order to administer virtual machines in a remote cloud. To my surprise however it was quite the opposite experience for me. When I logged in for the first time the interface was very clean and only showed me the basic options that I needed. Below is a screenshot of how easy it is to create a new virtual machine in vCloud Air, complete with a cost calculator.

Create a new VM in vCloud Air

Below you can see all of your resource usage across your own cloud. You can even get the same view on individual virtual machines with up to the latest hour worth of data.


I am a long time user of VMware Workstation and with the release of Workstation 11, the is also support for vCloud Air. So for those of you that do not administer VMware ESX hosts/clusters you can still fire up virtual machines in the cloud. It surprises me how many IT professionals are still building home labs when they can be utilizing on demand cloud resources. You can build a hybrid cloud now from a laptop with VMware workstation and vCloud Air OnDemand. Below are some screenshots of the workstation integration:

workstation 11 vcloud air

vmware vcloud air workstation
Besides the native integrations and ease of use lets talk a bit more about one of the easiest use cases for vCloud Air… Disaster Recovery. Late adopters of cloud technologies can’t argue the fact that DR is still a great use case for cloud technologies. Some of the newer features released include:

  • Native failback support – Customers will now be able to resume normal data center operations in their primary data center following a failover to vCloud Air. Customers will be able to replicate workloads back from vCloud Air to their primary environment over the network and resume normal operations.
  • Multiple recovery points – Customers will have the option to roll back to multiple earlier snapshots of their data center environment. This is vital to recover from outages caused by data corruption, viruses or hacking attacks that compromise the most recent recovery point.

If you haven’t already, go ahead and check it out. VMware is currently offering $300 in service credits for your first 90 days which is plenty of time to try out all of the features. As always if you have any questions or need me to elaborate on anything please comment below or follow me on twitter.



vmware CBT changed block tracking

VMware CBT may not be giving your backups accurate data.

If you are currently using VMware vSphere and you are utilizing CBT (Change Block Tracking) for you backups you may be at risk of bad backups. Luckily this issue only effects vmdk files that have grown over 128GB in size, so this should narrow down your scope of affected backups. This issue can be reproduced by expanding a virtual disk (vmdk) file with Change Block Tracking (CBT) enabled. Your virtual machine may be at risk if the vmdk file was extended to a size greater than 128 G. Which would incorrectly calculate the in-use blocks in the vmdk file.

This affects esxi hosts versions 4.x and 5.x and there is no currently no fix or update; There is however a workaround.


Go to each of the affected VMs and disable and then re-enable Change Block Tracking (CBT) on the virtual machines.

How to Disable CBT:

  1. Power off the virtual machine.
  2. Right-click the virtual machine and click Edit Settings.
  3. Click the Options tab.
  4. Click General under the Advanced section and then click Configuration Parameters. The Configuration Parameters dialog opens.
  5. Set the ctkEnabled parameter to false for the corresponding SCSI disk.

How to Enable CBT

Note: Ensure that you power off the virtual machine before performing these steps.
  1. Right-click the virtual machine and click Edit Settings.
  2. Click the Options tab.
  3. Click General under the Advanced section and then click Configuration Parameters. The Configuration Parameters dialog opens.
  4. Click Add Row.
  5. Add thectkEnabled parameter and then set its value to true.
  6. Click Add Row, add scsi0:0.ctkEnabled, and set its value to true.
    Note: scsi0:0 in scsi0:0.ctkEnabled indicates the SCSI device assigned to the hard disk that is added to the virtual machine. Every hard disk added to the virtual machine is given a SCSI device that appears similar to scsi0:0, scsi0:1, or scsi 1:1.
  7. In the home directory of the virtual machine, verify that each hard disk contains a vmname-ctk.vmdk file.



VMware’s Hyperconverged Infrastructure Appliance: EVO Rail

evo rail


At VMworld 2014 this year, a new hyperconverged infrastructure appliance emerged. This appliance brings together storage, networking and compute for an easy to manage IT infrastructure. A simple and powerful appliance taking a large step in the SDDC direction. When you no longer have to manage hardware, you can focus more on applications and data.

I put together this page as a central point for as much EVO information that I can find. Let me know if you find something else you would like me to add to this page, and I will continue to add information as I find it.

Official Product Page:

EVO Rail Datasheet



Marco Broeken
Julian Wood
Chris Wahl
Timothy Prickett Morgan
Louis Cheng


Project “MARVIN” stands for: Modular Automated Rackable Virtual Infrastructure Node.

OpenStack Icehouse Released Today

The latest version of OpenStack called Icehouse was released today.

New Features in Icehouse

OpenStack Icehouse has nearly 350 new features that reflect a community-wide effort to bring the voice of the user into the rapidly maturing open source cloud software platform. Key themes this release cycle have been raising the bar on testing and stability, as well as tighter integration across the platform. OpenStack Database Service (Trove) also became part of the integrated release in Icehouse.

Compute (Nova)

New support for rolling upgrades minimizes the impact to running workloads during the upgrade process. Testing requirements for third-party drivers have become more stringent, and scheduler performance is improved. Other enhancements include improved boot process reliability across platform services, new features exposed to end users via API updates (e.g., target machines by affinity) and more efficient access to the data layer to improve performance, especially at scale.

Object Storage (Swift)

Object Storage released discoverability, which dramatically improves workflows and saves time by allowing users to ask any Object Storage cloud what capabilities are available via API call. Additional improvements included a new replication process that improves performance with data that is more efficiently transported through s-sync instead of r-sync protocol. Significant progress was also made on ongoing storage policies features and will be further developed for the Juno release.

Block Storage (Cinder)

Block Storage added backend migrations with tiered storage environments, allowing for performance management in heterogeneous environments. Mandatory testing for external drivers now ensures a consistent user experience across storage platforms, and fully distributed services improve scalability. Overall quality, dependability and upgrades were a focus for Icehouse including fixing more than 400 bugs.

Networking (Neutron)

Networking introduced tighter integration with Compute in Icehouse that includes improved consistency with bulk instance creation and improved provisioning. There is better functional testing for actions that require coordination between multiple services, and third-party driver testing now ensures consistency and reliability across network implementations. Significant progress was made during the Icehouse release cycle to bring OpenStack Networking (Neutron) to feature parity with Nova-Networking, but both options are still available for users in the Icehouse release.

Dashboard (Horizon)

OpenStack Dashboard again expanded its multi-lingual support and now includes 16 languages with the addition of German, Hindi and Serbian. Additional improvements to the general user interface and experience were also implemented such as a more self-guided navigation, inline table editing and extensibility.

Identity Service (Keystone)

The first iteration of federated authentication is now supported allowing users to access private and public OpenStack clouds with the same credentials. General stability and performance improvements were also made.

Orchestration (Heat)

Automated scaling of additional resources across the platform, including compute, storage and networking is now available. A new configuration API brings more lifecycle management for applications, and new capabilities are available to end-users that were previously limited to cloud administrators. Collaboration with OASIS resulted in the TOSCA Simple Profile in YAML v1.0, demonstrating how the feedback and expertise of hands-on OpenStack developers can dramatically improve the applicability of standards.

Telemetry (Ceilometer)

Telemetry features improved access to metering data used for automated actions or billing / chargeback purposes.

Database Service (Trove)

A new capability included in the integrated release allows users to manage relational database services in an OpenStack environment.

Image Service (Glance)

The Image Service features improved calculation of storage quotas, and the project moved to using 0-based indices for location entries.

Google Data Center

Google’s Data Center

I’ve been in several of the largest and most secure data centers in the United States and it just occurred to me that as often as I use Google, I have never seen any of their data centers. I thought it would be fun to do some quick research and post about what I find here.

So where do we start ? of course ! It appears that Google has created a dedicated page to outline features in their data centers.

Roundtower Wins Ernst and Young Entrepreneur of the year Award

Ernst & Young’s Entrepreneur Of The Year is the world’s most prestigious business award for entrepreneurs. The unique award recognizes the significant contributions of entrepreneurs who inspire others with their vision, leadership and achievement.

I made a cameo in the video:

Official link to Press Release:


1 2 3 4