How to address the two key challenges with virtualization
Three papers for the data centre
This week we have curated three papers from the Reg Whitepaper Collection that concern themselves with various aspects of server and data centre performance.
This paper from Riverbed calls for companies to implement Wide Area Data Services (WDS) technology to help eliminate the performance penalties associated with distributed environments and run more appropriate disaster recovery strategies for server virtualisation shops.
Since the 1990s server virtualisation and consolidation have evolved to a point where they have become almost synonymous, Riverbed argues. But the benefits can come at a high price, with application performance deteriorating across disparate locations. The most common complaint network managers hear is that the network is “slow” or the application is “broken.” The client-server applications that worked smoothly on the LAN break down over the WAN and perform poorly, or not at all, and tasks that were instantaneous on the LAN now might take several minutes.
The main causes of poor application performance over the WAN fall into three main categories: bandwidth congestion, transport layer chattiness and latency, and application protocol inefficiencies. The paper runs through the issues really rather well and at the end presents an overview of Riverbed’s WDS offering.
Double-Take Software kicks off with advice on how to devise a strategy for virtual machine recovery. IT quickly moves on to the issues and headaches of back-up and Microsoft’s Hyper V virtualisation software. According to Double-Take, Hyper-V failover clustering has "management challenges that require skilled and experienced system administrators to avoid self-inflicted downtime”. The paper ends with a fairly techie discussion of Double-Take for Hyper-V, a high availability back-up solution The subject matter is a little dry, but plenty of Reg readers are downloading this paper.
Based on a survey of 164 IT execs, this IDC research note shoehorns data centre automation and virtualisation into the same pages. The executive summary is this: "While virtualisation continues to spread across the enterprise, process standardization and automation must be utilized to improve IT’s business impact and service delivery efficiency."
HP is the sponsor of this paper: automation is a key business and technology plank for the company. But there is no sales pitch, and plenty of interesting slide action, for comparison purposes.
Sponsored: RAID: End of an era?