I notice there are couple of significant improvement from this new generation blade as below

1. 2 sockets base blade servers to provide up to 16 cores with Intel E5 chip by utilizing 1600Mhz DDR3 memory
2. It provide up to 384 GB of memory with 16 GB memory module
3. IO interface through put increase up to 40Gbps per blad

I had done the comparison to B230M2, the B200M3 seems to be a better choice for high density computing which require higher memory but considering to reduce the foot print of the machine.

Great release from Cisco which provide more option for us to consider moving forward in the future.

Related posts:

1. Cisco UCS added new Fabric Interconnect to the Product Line
2. Cisco UCS Rank #3 in Worldwide Blade Server Market
3. Cisco and VMware Submitted VXLAN for Open standard

Let’s talk a little bit more about what is VXLAN and why do we need that. As of today, DR solution are commonly deploy by every enterprise environment. Did you realize 1 of the biggest challenge which always happen on the DNS record, hard coded IP segment, subnet and VLAN which always require human interaction to manually change it, to allow a success fail over of the DR scenario? Well, I think this is just 1 of the major issue commonly face. With Cisco Nexus 7000 come with OTV (Overlay Transport Virtualization) which allow you to stretch the Layer 2 traffic over Layer 3, with minimal bandwidth requirement. Cisco had done a great job to the enterprise to address this, which now you can have both side learning the same VLAN, Subnet and ip address. A DNS and ip changes is no longer require as both end are transparent with similar VLAN.

What about Cloud? I always think the major challenges on cloud are Network. VXLAN created are mainly to address to concerns of the public cloud provider. That was mainly due to the limitation on the current VLAN architecture, which only allow 4000++ VLAN to be created in the single manage environment. Imagine you as the public cloud provider, you will easily hit the maximum number and you do not want to build another new infrastructure due to this limit too. According to the Cisco Expert during VMworld, the additional header is been added to the bit and byte to enhance the additional number of VLAN could be deploy in the single environment. At the same time, it deliver the similarity of OTV from private to public cloud, which bring the Hybrid Cloud to the reality of tomorrow.

I will try to find out more technical information about this area with my counterpart back to Malaysia, and share again in details with another post. From the intention of both Cisco and VMware, they are pushing hard for VXLAN on this and I do believe this is 1 of the key area need to be addressed before the hybrid cloud environment could continue scale especially on the public cloud provider perspectives.

You can view the full details of the IETF submission from Cisco and VMware from HERE.

Especially on virtual desktop environment, workload offload is a must to be considered move forward to improve the consolidation ratio and maximize the the performance out of the infrastructure build. If you had not considered about this, you should immediately look into this as security can be another major bottlenecks to your infrastructure today, and of courses all the business units are always concern about it.

10Gbps Ethernet no longer as a dream to most enterprise environment. It had become affordable, and realistic to be adopted while users decided to build the virtualization or private enterprise cloud. I had always highlighted that network had become major component of the success for cloud or virtualization, therefore the manageability on the virtual switch had become equally important as well. Previously, Cisco Nexus 1000V is the only choice to perform virtual switch layer management. If you do not want to go for it, you have another choice here. Arista had a similar offer which is vEOS, that allow virtual traffic direct integrate and visible over to the Arista Physical Network switches.

Below are some table comparison within vEOS VS Cisco Nexus 1000V. There are some similarity but will not be the total same approach been taken.

Here is the table shown that some details functionality Compare to Cisco Nexus 1000V.

If you are planning for large scale deployment on 10G environment for production network and IP base storage network, you may also consider Arista as a good fit which they offer low latency and work well with most IP base storage vendor in the market. For more information, please refer to the official link here.

According to the link here www.theregister.co.uk,  it should able to push the total memory up to 1TB per 2 U machine in single chassis. The C260M2 will come with 2 x 10Gb Ethernet and 4 x gigabits NICs, which is very sufficient for most of the deployment today on stand alone or virtualize environment. Cisco claimed it will ship the B230 M2, B440M2 with Intel E-7 base chip set. Once again, I am very excited for this to reach the market as it will transform the server infrastructure and virtualization infrastructure deployment in the future.

You can read more from the original post here.

I found some very good post over the internet, you can refer the URL below for more information about the new update provided in Firmware 1.4

Unified Computing Blog

M. Sean McGee

Well, there are many users, partner, or even principle who are reluctant to accept changes or new thing, which claimed nexus 1000V doesn’t bring any value while created problems to their day to day job. The main reason behind as they struggling to manage and configured nexus 1000v, mostly because they are just not the right person to manage it. Please do not feel offended with this statement, as I am telling the truth which myself are not the right person to configure SAP, Oracle, BI cognos too. We always need specified specialist and expertise from different area to compliment each others in IT environment.

That is why Cisco and VMware are alligned to create this virtual switch Nexus 1000V, which Xen open source are trying to follow the same now. For Enterprise IT, proper change management and ownership are important to be in placed, unless you think that ITIL practices doesn’t bring any value to the IT operation. Without Nexus 1000V, Network operation are out of picture for the virtualization operation practices and support as there had no visibility due to NIC teaming, load balancing configurations and non-transparency and integration with security devices allow by using the DEFAULT DVSwtich or Standard vSwtich from VMware. This happen to all the previous Generation Virtualization deployed before vSphere 4.

For the story I heard, he blame the Nexus 1000V doesn’t work just because he struggling to create port profile, VLAN assignment and etc on the Cisco Nexus 1000V. This could be easily done by a network admin in any organization and a great team work and participation could be created while introducing Virtualization or Cloud computing in Enterprise Environment.

I agreed as not everyone will able to handle server, storage, security, network and facility in a single person. I will always advise my friends and buddies to understand the different functional team’s technology and operation which allow us to better understand the integration requirements from different pieces in the infrastructure. I am not a network expert, but I did pickup nexus 1000V as we learn from previous experience, without nexus 1000V could be a nightmare for large virtualization environment. If you are deploying a small scale virtualization for SMB with less than 10 VMs, you may not require Nexus 1000V, but if large scale deployment, please consider this solution as part of your technical design consideration.

Virtualization & Cloud Computing require Network to deliver the services by resources sharing, if security and manageability are not address properly, the risk to business is very high. “The world is transforming, Changes are necessary”, we will be either ON the train or OFF the train in the fast pace moving IT world.

The traditional vswitch or dvswitch did not provide control, manage as the physical switching environment in the data center. It allow administrator to provision new virtual machine and connect to network without proper request and change management in place. Let’s imagine, the effect will be the same as unauthorized persons connected a server to the production LAN without a change approval. In physical environment, we can shut off all the unused network ports in data center, in virtual, you will not able to perform this while using vswitch or dvswitch from vmware. Technology limitation for the past previous years does not allow network operation to be fully integrated until the Nexus 1000V from cisco & VMware join develop products come to place. This will allow administrator to manage and control the access level on the virtual environment as it does in physical switch. Transparency and Manageability is possible on virtual layer now.  Network performance analysis and troubleshooting become easier compare to previous.

vCloud director come to place to provide private & public cloud to enterprise. Many of the new features are great and useful to the cloud environment. Introducing virtual firewall, security control and others function in vcloud director, does help to improve the cloud infrastructure, but should this fall under the vmadmin , network admin or security admin roles? Users may need to reconsider the approach by introducing all this new technologies to the production environment. They may need to get the right expertise and team to manage each different functional. This will allow the IT organization to move forward together without breaching the rules & regulation been set in the infrastructure.

Do not panic as you can easily fix this by manually configure your extension key by using the command below

n1000v(config)# vmware vc extension-key

This should able to make your new VSM to have the same extension key as previous. Once you had done this, you should able to reconfigure your new VSM to have the same name with previous. The orphane nexus 1000V DV Switch will not happen in this case.
Good Luck!!!!!

Cisco Nexus 1000V is a enhance virtual distributed virtual switch which integrate to the VMware vSphere 4. With Cisco Nexus 1000V, you will contain VSM and VEM, which VSM will be the virtual appliances that host the management of Nexus 1000V, and VEM will be installed on each ESX host that been connected to the Nexus 1000V.  I will not go in too details about Nexus 1000V in this post as my intention is to clear up the confusion on the VN-Link meant by Cisco.

6 months ago, this new term had come to my career while someone had claimed that the VN-Link will allow virtual machine to direct interact to the physical IO devices on the PCI-E without hypervisor in place. Well, this features had been included in VMware for quite some times in vSphere 4 by using the functionality call VMDirectPath I/O. But while doing this, you will lose the functionality of VMotion, and most users will not go for this due to the limitation on VMotion. Since this is new, no right or wrong, I decided to leave it for a while and further research to understand it.

After many round of discussion and reading on the blog and white paper, finally I have the answer I want now. VN-Link is the product name for a family of products, does not specifically refer to any one product so forget the idea of hardware vs. software implementation, etc.  Think of the Nexus family of switches: 1000v, 2000, 4000, 5000, 7000.  All different products solving different design goals but are components of the Data Center 3.0 portfolio from Cisco.  In VN-Link it contains Cisco Nexus 1000V, Hypervisor Bypass with Direct Path I/O, Pass Through Switching, Virtual Interface card & VNTag. If you had done the administration or implementation on Cisco UCS, Nexus 1000V, Nexus 5000 & 2000, you should familar with all these term that I mentioned here.

For more details information, I will recommend the Unified Computing Blog who had done the good explanation to described further in details for the VN-Link technology.