Haseeb Niazi is a technical marketing engineer in the Data Center Group specializing in security and data center technologies. His areas of expertise also include VPN and security, the Cisco Nexus product line, and FlexPod. Prior to joining the Data Center Group, he worked as a technical leader in the Solution Development Unit and as a solutions architect in Advanced Services. Haseeb holds a master of science degree in computer engineering from the University of Southern California. He’s CCIE certified (number 7848) and has 14 years of industry experience.
The following experts were helping Haseeb to answer few of the questions asked during the session: Derek Huckaby and David Klem. Derek and David are working with Haseeb and are experts in the field.
You can download the slides of the presentation in PDF format here. The related Ask The Expert sessions is available here. The Complete Recording of this live Webcast can be accessed here.
Q. Will Multisite FlexPod architecture support the VMware Site Recovery Manager (SRM) Release 5.x for disaster recovery?
A. SRM is not targeted for multisite solutions at the moment. The idea here is more of disaster avoidance than disaster recovery. Cisco wants to use the metro cluster and have the ability to move the workload from one site to another site. Although it is not to say that if you use SRM it will not work, but it has not been validated.
Q. Is MDS 9148 required for Multisite FlexPod Cluster?
A. Yes, the MDS 9148 is required for the back-end MetroCluster configuration. NetApp has validated Multisite cluster with MDS 9148. MDS switches are used here for back-end connectivity. The front-end can be Fibre Channel (FC), Fibre Channel over Ethernet (FCoE), Small Computer System Interface over IP (iSCSI) or any protocol you prefer.
Q. Are there any design solutions to implement firewalling in Flexpod to obtain Active|Active datacenters with L2 extension?
A. An update to the Secure Multi-Tenancy solution, which includes a firewalling component, is in process.
Q. Are there any differences in FlexPod and Vblock solutions except storage vendors? Do both use Unified Computing System (UCS), Nexus 5000, and VMware?
A. The idea and the concept is very much the same. Vblock is little bit more stringent, as you need to have a defined set of configuration (hardware and software). On the other hand, FlexPod is more flexible as it lets you choose. Also, Cisco provides a validation guide based on your input. So generally the method in which the problem is approached is slightly different.
Q. Do you have complete logical and physical building guides for FlexPod like VCE has for Vblock?
A. Yes, all of the logical and physical building guides are documented in our Cisco Validated Designs (CVDs). Visit http://www.cisco.com/go/flexpod to download the latest copies.
Q. Can you provide a document that explains the support services for FlexPod? For example, minimal service levels, stock-keeping units (SKUs), a customer interaction algorithm, and so on?
A. It stands for Serial Attached SCSI. Basically, it is how Cisco connects the NetApp controllers to the disks that are used. (As opposed to FC or
Serial Advance Technology Attachment (SATA) connections).
Q. In a really small installation, can the servers go directly into the FI and skip the 2232s?
A. While this is a supported configuration by both Cisco and NetApp, it is not valid within FlexPod today due to certain redundancy and scalability trade-offs made for that configuration.
Q. Is there a distance limitation for Clustered Data ONTAP?
A. Yes, clustered Data ONTAP supports lengths of up to 300m, which is the limitation of 10GbE cabling today.
Q. What application is used for 5548 FC zoning (Data Center Network Manager)?
A. In our CVDs, we use the CLI in order to configure all FC zoning within the Nexus 5000.
Q. What is the performance improvement over Nexus connectivity with FI and NetApp with FCoE compared to FC connectivity?
A. There has not been a specific study completed on the infrastructure side. However, a known improvement is that FCoE links are faster (10 G links compared to 4 G links). Cisco does not have specific information around performance with each protocol.
Q. How does vMotion occur if the datastore does not move quickly?
A. If you follow the VMware vSphere Metro Storage Cluster (vMSC) design guideline, you do not want to use vMotion. You want to define VMs and tie them to a particular data center. However, in cases where you want to perform vMotion, your VM gets migrated and continues to access its disk over the WAN link which is not ideal. So for that duration VM works, but the traffic moves back and forth between one data center to another data center. You can start the storage vMotion and move your disks. Once it is complete it will be available locally in that data center.
Q. Are there any restrictions/issues with the use of Violin instead of NetApp?
A. NetApp is an integral component of FlexPod. All the validation is completed in NetApp. Cisco has not completed any validation with Violin or any other storage vendor.
Q. Will Unified Computing System (UCS) Director provide a ready configuration for FlexPod based on CVD, such as a Day 1 configuration for quick configuration?
A. This configuration is being worked on by the FlexPod and UCS Director teams. Day 1 and Day 2 configurations will be implemented in future releases. A Day 0 configuration can be completed as well. Therefore, UCS Director will support Day 0 and Day x,y,z configurations.