The Riverbed Blog (testing)

A blog in search of a tagline

Upgrading Performance From Good to Great

Posted by riverbedtest on June 24, 2011

As a sales engineer, clients commonly ask me ‘what else could we do to improve application performance over the WAN?’ I normally respond with something like, “Keeping the network available and using WAN acceleration to make applications faster provides good performance. However, integrated application-layer Quality of Service (QoS) and robust performance visibility provides GREAT performance.”  QoS enables predictable performance over the WAN. Robust performance visibility provides the ability to efficiently solve problems.  If you want a worry-free WAN, in which business applications run well even under the heaviest utilization loads, then today’s blog post is for you.

Earlier this year, RiOS 6.5 was released with a major QoS overhaul.  This included application-layer classification via AppFlow; basic/advanced modes with robust traffic shaping improvements; and workflow improvements to save tons of time. If you want to ensure your precious WAN resources are mapped to strategic priorities, then RiOS 6.5 is your new best friend. With it customers can easily control tricky recreational applications, enterprise port hopping applications like Microsoft Exchange/Outlook’s RPC, and UDP-based Voice/Video.  This is a huge benefit for triple-play networks where balancing the simultaneous demand of voice, video, and data can be daunting. While QoS isn’t sexy, it enables customers to ensure performance when the WAN is most heavily utilized…instead of getting a call complaining about application problems just as they’re sitting down for dinner with the family. 

Blog_osi_modelWhile customers could use router-based QoS, they’ll find deploying WAN Acceleration and QoS in two different devices is less than ideal. Application-layer classification is critical in today’s networks where many applications tunnel over HTTP or port hop (Outlook/Exchange, VoIP, etc). Applying application layer QoS is the biggest challenge for enterprises since the WAN Accelerator obscures OSI Layers 5 – 7 (the application signatures) from the router. If you want application layer classification then moving QoS to the Steelhead is the best option. The Steelhead can see all the application signatures prior to optimization and apply the most appropriate QoS level. If the WAN supports Classes of Service then the Steelhead can also intelligently mark DiffServ bits, so an application’s packets are treated accordingly across the whole network. This enables an application-aware end-to-end QoS  architecture. To learn how to configure the new QoS features in RiOS 6.5 refer to Chapter 3 of the RiOS 6.5 Riverbed Deployment Guide and Chapter 7 of the Steelhead Management Console User Guide.

Blog_dashboardWith the acquisitions of Mazu and CACE Technologies Riverbed gained several unique visibility solutions…Cascade Profiler, Pilot, and Shark. The rapid integration of these solutions with Steelheads has delivered a formidable and cost-effective performance analysis toolset that tightly integrates with WAN optimization as a whole. Steelheads have long supported NetFlow export and tcpdump for packet captures. With Steelhead’s adoption of cascade-flow export, organizations can migrate from a basic utilization-focused view with Netflow to cascade-flow’s performance-centric view. This enables streamlined monitoring/troubleshooting for applications, sites, networks and users. The customers I’ve had deploy Cascade absolutely love the service dashboards, Active Directory user tracking integration, and true application-layer visibility. When they first migrate from the port and IP address visibility of NetFlow to the application visibility of Cascade they are amazed at what applications are really on their networks.

Blog_pilotIf they layer on Pilot and Shark they get high-speed packet capture, transaction visualizations, and packet analysis to quickly find the root cause of problems. Riverbed has a long tradition helping customers set new IT speed records, and we’ve just done it again with the addition of Cascade Profiler’s right-click integration with Pilot/Shark. Customers now can drill-down from reports in Cascade Profiler to the transaction and packet visualizations in Pilot all in the same workflow.  If they need to go even deeper they just right-click and open only the packets they desire in Wireshark where they can see the exact bit and bytes. My customers have been particularly impressed with Pilot's ability to efficiently find and analyze VoIP call problems. Finding and analyzing performance problems has never been easier.  To see videos of  Pilot in action go here.

At the end of day, if an organization can’t proactively solve network, application and user performance challenges then they won’t have great performance. When WAN optimization is deployed across an enterprise proper planning and execution to ensure the IT staff has the right visibility into the right data is worth the effort.  With Steelheads at remote sites integrated with Cascade at the datacenter customers now have enterprise-wide visibility to efficiently find and analyze performance challenges.

Don’t settle for good performance, upgrade to great performance today.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: