Running Remote Sites? Lines chugging with data flow? Don’t need the overhead of a high speed, high cost leased line?
Then storage virtualisation with DataCore Software and WAN Optimisation with Riverbed could hold the answer: Our testing at a
DataCore’s new Disaster Recovery white paper reveals how organisations can reduce bandwidth and achieve more stringent Recovery Time Objectives (RTO) and Recovery Point Objectives (RPO) without incurring the expense of higher speed inter-site lines.
Working hand-in-hand with accredited partner Waterstons, DataCore and Riverbed have explored and demonstrated how the bottleneck of line capacity to remote and Disaster Recovery (DR) sites can be overcome without deploying high-speed, high-cost lines.
The Problem: Growth in virtual volume traffic from the Disaster Recovery site causes line capacity overload:
Growth in virtual volume traffic from the Disaster Recovery site causes line capacity overload Waterston’s noticed that as more organisations consolidated from physical to virtual servers, the inter-site traffic between the primary and DR site was growing so fast that the volume of updates from virtual machines exceeded the line’s transmission capabilities. This lag unacceptably exposed remote DR sites to data loss. An obvious solution was to upgrade to higher speed, higher capacity lines but this entails significant infrastructure overhaul and cost in recurring leased line rentals.
Alternative solution – remote replication within DataCore’s SANmelody + compressing, caching and optimisation by Riverbed Steelhead network appliances
The IT consultants at Waterstons and DataCore devised an alternative solution through a combination of storage virtualisation and WAN optimisation. They employed the built-in remote replication function in DataCore SANmelody™ software to relay changes in virtual disk blocks from the primary site to the DR site over an existing 2MBps IP WAN. These inter-site transmissions were compressed, cached and optimized inline by inserting Riverbed Steelhead network appliances on either side of the link. The solution succeeded in meeting the firm’s disaster recovery objectives and proved much more cost-effective than upgrading to higher bandwidth service.
Access a full copy of the white paper to find out more, including:
· How it works – remote replication via DataCore storage virtualisation software
· How WAN optimisation from Riverbed Steelhead network appliances eliminates up to 98% of application roundtrips
· Real-life scenario – a leading syndicate of salvage companies in the North of England verifies the Waterston solution
· Test results & explanations
Customer Test Conclusions & ROI
Through the live test analysis, Waterstons revealed that the combined solution can drop the time to remotely replicate three newly created VMs from 42 hours down to 3 hours, and reduce the ongoing replication window to four minutes – substantially improving the customers’ achievable Recovery Time Objective. The salvage company can now recover more than two day’s worth of data that may have been lost in their un-optimized configuration.
Furthermore, DataCore’s remote replication software allows customers to map snapshots of the remotely replicated volumes to the ESX hosts at the DR site to dramatically simplify and speed up recovery without painful back-ups andrestores.
Equally compelling is the cost savings afforded by these technologies. Compare the alternative average 2Mbps increased bandwidth combined rental and initial configuration of £22,000 – to upgrade the line by a factor of 2x – versus the total cost of the combined and DataCore/Riverbed solution of £16,800 – with an annual cost of £2,800 that improves the performance tenfold (10x).