(The Hosting News) – RackWare, the software provider that integrates data center and cloud resources into a scalable and intelligently managed computing environment, today announced top predictions for the year ahead in the cloud, particularly in cloud-based disaster recovery (DR), datacenter heterogeneity and workloads in the cloud. Based on RackWare’s observations, there has been significant progress made with cloud technologies in 2014, however there is plenty of room for advancement in 2015.
RackWare predicts that the following areas will be of particular importance in 2015:
1. Cloud-Based DR Will Close The Gap Between High Availability Software Solutions, And Tape / Image Archives
For mission-critical workloads, real-time replication or high availability (HA) solutions offer the best protection, but are typically complex and expensive. The advent of Cloud infrastructure has made it easy to take a “snapshot” of workloads with ease. Cloud DR is more dynamic than tape / images, but less expensive than HA solutions. The images are not static in the classical sense, in that they can be “turned-on” when needed. Since they don’t need to be restored to run, they eliminate the manual effort and errors that may occur during the restore process. Since public clouds are already located off-site, Cloud DR gives an added bonus of giving off-premises redundancy for traditional physical data centers. The concept of auto-failover and failback will be transferred from traditional complex HA systems into Cloud DR, making Cloud DR a cost effective bridge between HA and tape backup.
2. Cloud DR Will Increasingly Be Used For Planned Outages
Cloud DR can be used to address planned downtime just as much as unplanned downtime. Disasters are quite rare, but planned outages are very common — these include batch processing, reporting windows, and of course backup windows. Instead of consuming resources from production workloads, the synchronized “clone” image of the original workload can be started in the Cloud and used for processing. Cloud-based hot backups will be used increasingly for offline batch processing, and reporting, decreasing impacts on production workloads, and reducing and eliminating offline-processing windows by off-loading all processing to the Cloud.
3. Cloud-Based Hot Backups Will Be Used Increasingly For Test/Dev
Some data centers clone their production environment fully or partially in order to perform applications testing before promotion into a live environment. When application changes are made, they are placed in the replicated data center until they are fully tested, and then the changes are promoted into production, usually through another set of staging servers. Maintaining a second, and sometimes a third, identical data center is expensive and resource intensive. By utilizing the failover and failback facilities of Cloud DR, production hardware no longer needs to be duplicated. Entire data centers can be replicated into a cloud environment and used as the test lab for staging and promotion into production.
4. Disaster Recovery As A Service Will Gain Momentum
The concept of self-service will emerge in DR where end users can automate backups without assistance from IT professionals. We will see a tendency to move away from expensive, traditional DR to Recovery as a Service in the Cloud. Application owners will have the ability to setup DR strategies themselves, freeing time from IT admin and operations. With the help of third party solutions, MSPs, ISPs and cloud providers will setup services to automate all aspects of a DR strategy for their customers. Features that were once available only to mission critical workloads in a traditional data center will be available as a service to any data center residing in any kind of infrastructure.
5. Clouds Will Increase Heterogeneity In The Data Center
Many enterprises are diversifying away from, or moving completely over to new cloud infrastructure. This is occurring primarily due to anticipated cost savings from the pay-per-use model of the public cloud, where companies only pay for the bandwidth, processing, storage and memory resources that are used per unit of time. We have also observed a trend towards “cloud-consolidation” where enterprises are consolidating to new clouds from old clouds or to public clouds from private clouds. With all these new infrastructure options, data centers are hosting an increasing number of infrastructure types: public, private, virtual, physical across different operating systems, hardware and cloud vendors.
6. Legacy Applications Will Continue To Persist In The Cloud
Today’s leading edge application is tomorrow’s legacy application. People processes and operational dependencies are reasons why applications remain in the data center and never go away, even when they are very old. Interestingly, it appears cloud is not only for green-field apps, as many legacy apps are being moved to the cloud. We have observed a trend towards stitching together unchanged internal infrastructure with cloud resources, rather than re-architecting internal infrastructure into private clouds first. In other words, some data centers are migrating their legacy applications directly into the public cloud and bypassing the private cloud altogether. Some data centers are even moving back from public cloud into their own traditional data center.
7. Mergers And Acquisitions Will Continue To Increase Heterogeneity
M&A events compound the data center infrastructure heterogeneity issue even further. In a company merger, one company may be in a private cloud, while the other company has a mixed traditional data center and public cloud environment. There may be redundant workloads, or the applications might need to be integrated intimately. Since heterogeneous data centers are never going away, there will be a great need for tools and solutions that can move workloads freely between environments, whether they are public, private, or traditional data centers, in all possible directions.
8. Enterprises Will Become More Serious About Running Crucial Workloads In The Cloud
Public clouds have gone through enough upheavals over the years and have become hardened and made to withstand constant hacking attempts. Public clouds by their very nature are accessible by anyone on the Internet, and security has been made extremely robust as a result, time-tested through countless hacking attempts across industries, companies and individuals. Arguably, the public cloud has had more time exposure to potential security upheavals than its private cloud, or traditional data center counterparts. We anticipate that public cloud will become even more robust over time, as it appears that some data centers have come to recognize that public clouds may actually be more secure than internally built environments.
9. Tools Will Be Needed To Move Workloads From Where They Are To Where They Need To Be
Over the longer term, workloads will automatically move themselves in order to balance capacity demand with resource supply. The concept of self-service migration will take hold in data center environments for workload portability, allowing users to move workloads between dissimilar infrastructure (physical, virtual, public, or private cloud). Users will be able to move their own workloads from where they are, to where they need to be, no matter what infrastructure the workloads are running on. Application owners and end users will migrate their own workloads, without assistance from IT. Over the longer term, workloads will automatically scale out into the cloud when needed, and scale back in when the resources are not needed, all based on elastic demand and supply — heterogeneous, cross infrastructure tools will make this possible by third-party software vendors — and these tools will likely not be made available by the cloud vendors themselves.
With these in mind, 2015 is shaping up to be a year of remarkable innovation and progress for cloud technologies. RackWare looks forward to continuing to push the boundaries in the New Year and promote cloud advances.
RackWare allows enterprises to use the public cloud as just another resource for their internal infrastructure — for disaster recovery, as well as scaling purposes. With its unique ability to be platform and cloud agnostic, RackWare’s flagship solution, the RackWare Management Module (RMM), allows workloads to be ported between any platform, virtual or physical, and any cloud. RackWare has moved thousands of workloads for hundreds of customers and has partnerships with large Service Providers and VARs. RackWare was founded in 2009 and is based in Milpitas, Calif. For more information, go to: www.rackwareinc.com.