As its business grows, Lufthansa Systems wants to anticipate customer needs for high-availability and disaster-recovery solutions. To meet this goal, it plans to adopt Microsoft SQL Server 2014 and Windows Azure. In a successful pilot, the Microsoft
technology led to even faster and fuller data recovery, reduced costs, and the potential for a vastly increased focus on customer service and solutions, compared with the company’s current solutions.
A data center—any data center—is only as good as its disaster-recovery system.
Lufthansa Systems AG understands this well. The full-spectrum IT consulting and services organization serves not only its parent airline, but also other major airlines, plus airports, financial services firms, healthcare systems, publishers, logistics companies,
among other businesses.
To do so, it operates one of Europe’s most powerful data centers—in Kelsterbach, Germany, in addition to data centers in the United Kingdom, the United States, and Singapore. Those Lufthansa Systems data centers host Microsoft server applications, third-party
commercial applications, and custom and line-of-business applications.
To provide high availability and disaster recovery to its parent and its other customers, Lufthansa Systems operates failover clusters and database log shipping. At a minimum, that requires primary and backup data centers at a given site. For customers with
the strictest requirements, it can require geographically dispersed sites—say, in Germany and the United Kingdom.
But in an environment in which every second of downtime can mean hundreds of thousands of dollars in lost revenues for Lufthansa Systems’ customers, the minutes involved in log shipping to a recovery site, and the minutes more in reestablishing a database,
aren’t always acceptable.
Yet, the more powerful the solution, the more expensive it is: more infrastructure must be bought although it will mostly sit idle, and more staff must be hired although they may sit idle, too. It’s far from cost-effective.
Lufthansa Systems wanted a high-availability and disaster-recovery solution that was faster, more reliable, and more cost-effective.
||We chose Microsoft technology because no one else can deliver a complete solution. We save the time, money, and hassles of integrating various pieces.
| Bardo Werum
Senior Vice President Infrastructure
The organization is currently piloting a high-availability and disaster-recovery solution using Windows Server 2012 operating system, Microsoft SQL Server 2014 software, and the Windows Azure cloud services. Executives expect to deploy the solution on a
rolling basis starting in 2014.
“We chose Microsoft technology because no one else can deliver a complete solution,” says Bardo Werum,, Senior Vice President Infrastructure at Lufthansa Systems. “We save the time, money, and hassles of integrating various pieces.”
The pilot project focuses on boosting availability and disaster recovery for a business-critical enterprise system that processes data from 4,000 instances of the Windows Server operating system and 400 instances of SQL Server.
The pilot’s architecture is a hybrid on-premises/cloud combination consisting of two virtual instances of SQL Server 2014 at the Kelsterbach data center and another virtual instance in Windows Azure, all of which are part of a coordinated SQL Server AlwaysOn
The two on-premises instances are a primary replica and a secondary replica operating synchronously to support high availability through automatic failover. The Windows Azure instance is another secondary replica, operating in asynchronous mode to provide
disaster recovery through manual failover.
Lufthansa Systems tests the architecture by importing all system data into it and repeatedly simulating both soft and hard shutdowns of single and multiple nodes, as well as restorations of the nodes. Primary failovers are to on-premises nodes, with the Windows
Azure nodes used in simulations of complete shutdowns of the on-premises nodes.
The tests also confirm the operation of Windows Server Clustering, on which the operation of the Availability Group depends. Lufthansa Systems is also successfully testing the use of replicas—when they were not needed for disaster recovery—for faster and
more frequent system reporting.
Lufthansa Systems anticipates using Microsoft high-availability and disaster-recovery technology for even faster and fuller data recovery, reduced costs and deployment times, and a keener focus on customer solutions, compared with its current solutions.
Recovery Time Cut from Minutes to Seconds
The Microsoft solution meets the organization’s needs for high availability and disaster recovery. “We’re seeing both lower RTOs [recovery time objectives] and better RPOs [recovery point objectives] with SQL Server 2014 and Windows Azure,” says Werum. “Recovery
times, for example, have dropped from minutes to seconds. And RPOs have dropped to near zero; we’re seeing almost no data loss.”
These are important benefits for business-critical applications. According to Werum, “The Microsoft solution will help us to provide better service to our customers—which in turn will help them to provide better service to their customers.”
Costs and Deployment Times Decrease
Lufthansa Systems also says that the Microsoft solution will deliver its superior results while reducing cost. “We expect that making Windows Azure part of our disaster-recovery solution will reduce costs significantly,” says Werum. “That makes us more cost-effective,
which makes us more competitive.”
Werum also envisions using Windows Azure to reduce the time to deploy disaster-recovery solutions for customers. “We can use Windows Azure to reduce the time to put a disaster recovery solution into production by days. Our customers want solutions deployed
as soon as possible; this will help us to do that.”
Move to Cloud Provides Option for a Major Increase in Customer Services
According to Werum, the hybrid solution is the first step in a potential full move to the cloud. “This is a way to test the public cloud and Microsoft technologies,” he says. “As contracts for our rented data center facilities expire, we’ll have the option
to move our business to the cloud. As fewer of our people need to manage infrastructure, they can focus on customer-specific services, BI, analysis, and application services. The move to Microsoft cloud services could mean a huge increase in our ability to
deliver customer solutions.”
This case study is for informational purposes only.