Here in Vegas for the VMware Partner Exchange event taking place this week, Feb 12-16th, this quote from Curtis Sahakian could not be more fitting. Whether it’s your luck at the poker table, or all the exercise you get walking around the convention center, or the opportunity to network amongst partners at this event – hopefully this motto rings some truth for you.
So let’s work together to increase your profitability and differentiate the value you provide to customers by including Brocade networking solutions, the best networking products for virtualization and cloud. Visit us at Brocade Booth #208 to learn about our compelling joint solutions that have been tested and validated with VMware:
Virtual End User Computing (VDI): The Brocade/VMware solution deploys VMware’s innovative virtual end user computing on the Brocade VDX family of data center switches and Ethernet fabric.
vMotion and Load Balancing across data centers: Brocade, VMware, and EMC have a reference architecture to implement a stretched vSphere cluster across two sites within a metro area.
on 10-26-201308:05 AM - last edited on 10-28-201301:13 PM by bcm1
In a recent VMware press release, CEO of VMware Paul Maritz says “Cloud computing represents the next major era of computing – and it offers major opportunities to simplify and enhance IT. Now is the time to harness the benefits of cloud computing to enhance IT’s ability to support mission goals and enable government to do more with less. And by beginning the cloud journey with virtualization, agencies can drive significant cost, agility and security benefits." Although Mr. Maritz is evangelizing the benefits of Cloud Computing (Private, Public and Hybrid Clouds) in the context of Government customers, IMHO, the benefits apply to other Businesses and Enterprises as well. Virtualization as a core technology enabling Cloud Computing is a disruptive technology and has brought down cost of solving problems of Business Continuity and Disaster Recovery.
Business continuity and 24x7 application availability are top requirements for any organization. Advances in virtualization, storage, and networking have made enhanced business continuity possible, so that IT can deal with unplanned outages (disaster recovery) as well as with planned disruptions (disaster avoidance) and eventually take advantage of the flexibility and cost efficiencies offered in the cloud. The disruptive virtualization technologies and products from VMware in working with partners such as EMC and Brocade are bringing down the cost of deploying business continuity solutions.
Brocade, EMC, and VMware have collaborated to test and validate a solution that offers IT Operations the capability to dynamically migrate applications live across data centers without business interruption. This demonstration of live application mobility across data centers separated by a latency of up to 10 ms round trip (~550 miles) relies on the industry-proven VMware® Metro vMotion™, with the EMC VPLEX Metro application mobility solution, running on Brocade® IP and storage networking infrastructure for high performance traffic management and DC interconnect. The workflow is automated by the Brocade Application Resource Broker to ensure a non-disruptive end-user experience throughout the vMotion event.
Building the right storage and network infrastructure to enable data and application mobility requires a data center infrastructure that can provide both optimal storage extension capabilities as well as advanced network functionality end-to-end. It is imperative that a complete, comprehensive solution comprising storage, network, and server infrastructure be designed and implemented to facilitate the movement of applications across data centers. These products from Brocade, EMC, and VMware have been validated in the joint solution:
VMware vSphere 5 with Enterprise Plus licensing to enable Metro vMotion
EMC VPLEX Metro to ensure the availability of access to the storage in both data centers
Brocade VDX™ switching platforms, built on Brocade VCS technology, to provide a scalable and resilient Ethernet fabric for VMware server attachment at the network edge in a top of rack deployment
Brocade MLX and CES platforms for IP network aggregation and core and L2 data center extension with MPLS based VPLS
Brocade DCX Directors for storage extension, with FCIP and FastWrite technology
Brocade Application Resource Broker (ARB)–working in tandem with the Brocade ADX -- to provide high availability and seamless client access to the application by combining Global Server Load Balancing (GSLB) with the automated application mobility capabilities of ARB.
Brocade ADX for application high availability and seamless client access to VMs
Brocade Converged Network Adaptors (CNAs) and Host Bus Adaptors (HBAs) for ESX/ESXi servers for Ethernet and fiber channel connectivity to the IP and storage networks, respectively
Brocade Network Advisor for unified end-to-end solution management, including IP and storage networks
The solution comprises of four distinct areas of technology integrations. The solution is architected as;
Use of EMC VPLEX Directors to provide Active/Active storage for data availability for vMotion to be successful. The Brocade DCX Director class switch with the extension blades (FX8-24) is used to extend the SAN over FCIP. The vSphere servers with the Brocade 1020 Converged Network Adapters (CNA) are used to provide access to the VMware VMFS data stores.
The vSphere migration network or the network on which the Virtual Machine (VM) is migrated is connected to the Wide Area Network (WAN) using the Brocade Virtual Cluster Switching (VCS) technology
The layer 2 extension providing the IP address compatibility for the VMs is enabled by the Brocade Ethernet Fabric provided by the Brocade VDX coupled with the standards-based MPLS VPLS (Virtual Private LAN Switching) technology of the Brocade MLX switches in the two data centers.
The client traffic management which is paramount to ensure that the end user perceives no disruption in service is provided by the use of Brocade ADX integrated with the Brocade Application Resource Broker (ARB). A vCenter plugin ensures that the clients are redirected automatically to the new location of the application.
SPECjbb emulates a 3-tier system, the most common type of server-side Java application today. Business logic and object manipulation, the work of the middle tier, predominate; clients are replaced by user threads, and database storage by Java collections. The benchmark steps through increasing amounts of work, providing a graphical view of scalability. The benchmark is a CLI oriented benchmark used to stress the migration network and migration process within the vSphere servers.
SwingBench is a load generator used to stress test the network IP management between data centers. The benchmark is primarily used to stress the working of the Application Resource Broker (ARB) and the Application Delivery Controller (ADX) of IP traffic management between data centers. As you can see from the graph below, there is no appreciable effect on the benchmark during the migration.
DVD Store Application Mobility Testing
This set of tests measured the orders per minute (OPM) for the DVD store application with different simulated WAN delay values. The VM was configured to move from the primary data center to the remote data center while clients were accessing the application. Delays of 3 and 5 ms were introduced and OPM was measured at 19,500 and 18,500, respectively. The 5 ms round trip time (RTT) was introduced to emulate latency between the two data centers, which is the currently supported value for VPLEX Metro.
Additional proof of concept tests were performed to go beyond the 5 ms value with an RTT of 7.5 ms and 10 ms; the results showed 16,500 OPM and 13,869 OPM, respectively, with no application disruption. The latency of 10 ms approximately translates into 1000 kilometers between the vSphere servers. With the assumption that the data center LAN network has 500 ms latency within the data center, it is then safe to assume a latency of 9 ms or a distance of 900 kilometers or 562.5 miles between data centers.
The VM was able to complete the full migration without any traffic disruption. Clients connected to the application did not see any interruption. The Brocade ADX redirected client sessions to the new VM in the remote data centers. The ADX, with the global server load balancing and ARB capabilities, ensured that new clients were able to transparently connect directly to the remote data center.
Measurements Performance Data
The Brocade FastWrite feature is designed to overcome the latency effects for write operations, without compromising data integrity and security. FastWrite allows the entire data sequence of a write operation to be transported across a link, without the inefficiencies of waiting for the “transfer ready” to travel back across the high-latency environment. Brocade FastWrite is available with either FC-based or FCIP extension.
In addition to the tests for application performance and migration time, additional tests were performed to show the positive effects of the Brocade FastWrite capability available on the DCX backbone directors. Sequential write operations at different block sizes and different delay values between the two data centers were performed using IOmeter. Tests results showed that response time and IOPS were improved more than 2.5 times compared to not using FastWrite.
Remember to watch the video demo of above solution at Brocade booth #710 in VMworld 2011 and discuss about the solution in-depth with our solution engineers.
I would like to thank Ravindra Neelkant of VMware for helping folks at Brocade in architecting and testing this solution.
on 10-26-201307:18 AM - last edited on 10-28-201301:15 PM by bcm1
Brocade is the acknowledged leader in Fibre Channel SAN storage networks for block storage. But, much of the growth in storage today is in file storage, or what is commonly known as NAS (Network Attached Storage). NAS storage offers a robust, scale-out, high performance storage layer that simplifies storage management. However, the underlying network infrastructure for NAS storage has to be designed with the highest levels of resiliency, availability and performance. A lossless Ethernet transport can improve performance when network congestion occurs. TCP relies on the TCP sliding window for flow control. But, if congestion causes frame loss, then TCP recovery can severely restrict bandwidth as thetransmit window size shrinks and then slowly increases again. Enter Brocade’sVCS Fabric technology and VDX switch family (VDX 6710, VDX 6720, VDX 6730and the newest VDX 8770 switch). By handling congestion at layer 2 with lossless Ethernet, a VCS Fabric can avoid TCP frame loss and the time consuming recovery of full TCP transmit window size.
For the same reasons SAN fabrics are the best practice for high performance, resiliency and lossless block IO with Fibre Channel, VCS Fabrics bring these same qualities to Ethernet making a VCS Fabric ideal to transport NAS storage traffic.
Scale-out NAS Publications
The Strategic Solutions Lab recently posted several design guides and a Scale-out NAS primer showing how to cost-effectively use VCS Fabrics with NAS storage pools.
Wayne Tull, a Principal Systems Architect at Brocade, put together a very nice demonstration video showing how VCS Fabric resiliency overcomes link, path and device failures with NAS storage traffic and VMware server virtualization. What’s very cool in this video is seeing how virtual server IO to NAS storage never stops no matter what kind of failure occurs in the data path. Take a look at the last demonstration where he fails the primary IO path while a live VM migration is going on. There is no detectable pause in migration and no loss of application processing. Imagine what could have happened in a less robust network if this kind of failure caused a corruption to a database taking many hours to recover.
It’s been a bit quiet on the fabric front lately, what with all the hoopla around SDN. But there are different kinds of quiet. There’s “nothing going on” quiet. There’s “lots of secret stuff going on” quiet. And there’s “just busy getting stuff done” quiet. With ourannouncement today, it should be pretty clear that as far as our customers are concerned, it’s the last one.
This isn’t dramatic “I’ve redone my entire data center!” stuff. That’s kind of the point, in fact. Fabrics don’t represent a grand departure from current norms in terms of skill sets, org charts, or other exciting and scary career-changing things. They just make life for operators a lot easier, and their networks a lot more flexible...