“Transparency” is definitely one of these words which describe very well the Cisco culture. This is pretty obvious for a Cisco employee, but I got many times positive feedbacks from customers on the willingness of the company to share with customers and partners, so that they can benefit from the experience of a large and innovative community. This is not different in the data center. Some other companies on the market may have a different understanding of the communication around their data centers in maintaining a great secret about strategy, architecture and operations, but at Cisco we strongly believe in the power of sharing knowledge. I guess also that we are not afraid of the competition learning about our choices, as we see ourselves as leaders and we do support actively and constantly standardization.
When at Cisco Live Barcelona at the end of January, I met Donald de Witte, Cisco IT Program Manager and Rich Gore, Cisco IT Member of Technical Staff, who were present on the booth to talk to the visitors about the Richardson Data Center and the opportunity to visit immediately the facility.
As a global and dynamic company, Cisco like most of the leading companies embarked on a consolidation and virtualization path several years ago, with a plan to move from 50 data centers to 3 pairs of data centers. The Richardson Data Center is one of these brand new data centers, where Cisco IT has been able to take advantage of the new networking and computing solutions provided by Cisco engineers and partners, such as Unified Computing and Unified Fabric.
So are you ready to meet the Cisco It team and visit the Richardson Data Center ?
This is a favorite question of mine (and a very contemporary topic), albeit asked in a different way; Why do you need layer 2 for VMotion. Those bearing the email signature of VCP, will be quick to show off their shiny new badge and tell you it is because of VMotion, because of a port group label, or some other VMware specific component. In fact, there is a more fundamental reason. It is called an ARP cache, IP Address, Default Gateway and DNS.
Remember how hosts find each other – with IP a host looks to the IP address for the destination host, but remember we are speaking about L2 or to make it simple for this explanation, Ethernet. This is accomplished by using the address resolution protocol (ARP) to bind a destination IP address with a destination MAC (because Ethernet is Ethernet and NOT IP – Pet peeve alert when someone refers to Ethernet as an IP network, but I digress). Of course, the source host compares the destination IP with its own IP and mask, which determines whether you send the request to the default gateway – which in turn will proxy ARP for the destination host with its own MAC address, or to ARP on the local subnet. To avoid a constant broadcast storm, clients cache this information for a period of time with the assumption that not much or nothing will change.
We all know that VMware based data center virtualization is here to stay and is quite rightly incorporated in enterprise and service provider application delivery infrastructure. Although many benefits are gained by migrating to a virtual data center, deploying a virtual data center infrastructure that solves your application delivery needs can be challenging. The most common challenge faced is related to application deployment i.e. how will administrators create virtual machines (VMs), deploy applications, provision traffic flows between virtual servers and Application Delivery Controllers (ADCs), and perform routine maintenance of applications in a timely manner to meet business objectives.
Today quite a few ADC vendors have provided fragmented solutions which increase complexity. To realize the real advantage of virtual application delivery requires the application services to be integrated within the management environment of the virtual infrastructure and the virtualized data center. For example, in VMware’s case the vCenter will need the capability to provide appropriate controls for provisioning new VMs to the ADC, present views of traffic flows, and tools for automating change control associated with the application delivery framework.
Today this challenge has been addressed by one such ADC and it simplifies delivery of applications, even if they are deployed in virtual environments. Why?.. Because VMware has the capability to share compute resources among application loads and this next generation ADC enables sharing of the network bandwidth among network loads and provides vCenter-integrated tools and features to simplify application delivery.
The Olympics give us a chance to marvel at and enjoy the performances of athletes who demonstrate their intense dedication to the pursuit of excellence and their determination in the face of fierce competition. The games are a worldwide event and with audiences in the home countries anxious to see their athletes’ performances national television stations also have a challenge that they must meet. Fuji TV of Japan is no exception. With a long history of Japanese participation in the winter Olympics and a large home audience to serve, they have set up shop in Vancouver. Making this trip can be a costly undertaking and given the distance back to Japan they needed help in keeping communications smooth and keeping costs down.
To communicate with the home office Fuji TV set up a dedicated line between the company’s data center in Odaiba, Japan and the Olympic Media Village in Vancouver, a distance of over 7500 km. This connection provides Intranet access enabling the on-site staff, including production and news employees, to access the head office file servers and use its systems. This long distance WAN access allows Fuji TV to avoid having to transport and install data and application servers at Vancouver and thus avoid the cost of additional on site staff to maintain these servers. By reducing staff the costs savings is considerable, including air travel, housing, transportation and food.
We recently launched a new service capability out of our Customer Advocacy teams that provides a collaborative medium between facilities and IT operations. For anyone who has tried to manage the design aspects of establishing a new data center infrastructure architecture, you know that some of the most painful moments come from “translating” between facilities and IT designs. We’ve chosen to develop proposed infrastructure architectures using Google SketchUp which is freeware and incredibly easy to use.
Traditionally the physical design has been built using applications like Visio, AutoCad and BIM. These applications are ideal if you need a very detailed blueprint to build against. However, they tend to have large file sizes and are only accessible to specially trained individuals. With SketchUp and the 3D Warehouse that Google maintains (thank you Google!) we are able to provide a 3D wiki of sorts that allows facilities and IT to work towards a reconciled design intent. Once design intent is understood, the design specification process is quicker and far more simple.