It’s no accident that forward-thinking companies are developing new roles for dedicated “compliance officers.”
“Regulatory compliance” is the latest call to action for CXOs. We hear about it all over the place. The 2002 Sarbanes-Oxley Act is front and center because it is forcing enterprises to implement strict document management and data archiving policies. Failure to achieve compliance carries consequences that can include fines or prison sentences. Sarbanes-Oxley is not the only thing that people are worried about. Freedom of Information Act access requirements and other government regulations are also concerns for corporations more regularly in the target of information requests.
So it’s no accident that forward-thinking companies are developing new roles for dedicated “compliance officers.”
But for the most part, this headache is falling in the laps of overburdened IT directors. And the sad reality is that many companies are still unclear about their legal responsibilities when it comes to compliance. Often, companies merely err on the side of archiving everything to be sure.
IT managers are the ones who bear the burden of this as they struggle to replicate data across the enterprise so that it can be stored and consumed by the different parties struggling to maintain compliance.
Beyond the need to find the right data replication solution, there is the critical issue of how this impacts WAN bandwidth consumption. Data replication is a massive bandwidth consumer. In some instances, firms are backing up and storing massive amounts of data that heretofore was relegated to disk drives on a local basis at best.
The problem is that not many enterprises think about storing information efficiently. In fact, people have gotten downright sloppy in their data handling that the storage is dealt with after the fact. E-mail is a huge hog of disc space and bandwidth (few employees really consider the network impact of attaching an entire digital camera picture set to an e-mail and send it to a bunch of their friends). With multimedia being added to PowerPoint, XML links to Microsoft Office documents, and other such additive content, file sizes are getting bigger and bigger, and there’s more of it too.
This has obvious implications on the access links and WAN connections for enterprises. IT managers are not stupid – they see the growth in traffic and plan access and long haul connections accordingly, and increasingly use the Internet for transport as well.
In many instances, the best thing for an enterprise is to not just buy more bandwidth, but to better use the bandwidth that’s there. WAN optimization boxes have been around for a long time (heck, we remember when the first packet-based voice compression devices were launched and welcomed by international network uses). You can get boxes from players like Peribit Networks, Expand Networks and Packeteer.
The burden of data replication to meet regulatory requirements is causing many IT managers to look at such bandwidth optimization boxes to stretch their access and network dollars as far as they can go. That’s good. This makes sense.
But lately, these boxes have had to become something more than an ability to just reduce bandwidth consumption – they’re also about reducing latency, supporting QoS and monitoring too, items critical for applications like VoIP. You see, if you apply some of the older strategies for bandwidth optimization to today’s network applications, there can be a mismatch if the right standards and right prioritizations are not well supported.
Take Multi-protocol Label Switching. MPLS has become accepted because of its ability to offer prioritized services available across a network infrastructure. When QoS services are available on an MPLS network, network-bound traffic needs to have specific markings for the MPLS devices to treat data from those applications preferentially.
A challenge can arise, however, when packets are marked with TOS or DSCP on the LAN side but need a different marking for traversing the MPLS network. Avaya’s VoIP traffic, for example, uses two protocols – for signaling and bearer traffic – that use different TOS values. For traversing an MPLS network, though, they need to be treated the same and require prioritized service, so they need to share one TOS marking. Most WAN optimization platforms can only maintain or rewrite a TOS or DSCP marking, but what’s needed is the ability to layer on an additional “wrapper” marking for use in the MPLS network that is then stripped off again when the traffic arrives on the far-side LAN. Peribit Networks, for instance, provides WAN optimization products called Sequence Reducers and Sequence Mirrors that provide this ability to transparently add markings just for use in the MPLS network.
These kinds of technical challenges demonstrate how carriers need to work with IT managers to ensure that the necessary features are in place on WAN optimization platforms as they integrate with an MPLS network. Bandwidth optimization should be planned as part of a carrier’s line-up of supported network elements if it’s not already. It’s hard to be pitching advanced services like VoIP and QoS-based MPLS at the same time that users are deploying older or less functional non-compatible bandwidth optimization devices. Users and carriers alike need to be careful that older WAN optimization options don’t limit growth into today’s next-gen services.
