HP is making its annual splurge of enterprise product announcements at the Discover conference in Las Vegas this week. The announcements fall into two main categories – converged infrastructure, (including converged storage, networking and Big Data), and hybrid cloud.
HP said that companies are increasingly looking for more joined-up solutions that simplify IT management and enable them to quickly adapt to market conditions. While HP has offered an extensive enterprise product set for some time, the company admits that it has not always presented it in a coherent way.
The company's new and updated products aim to give customers practical solutions to some of today's biggest challenges, such as Big Data, network management and complexity in the cloud, while enabling them to continuously develop and improve their systems.
“The days when every new technology meant a major upgrade of your systems have gone away,” said Mateen Greenway, Chief Technologist for HP Enterprise Services. “Much more it's now this pragmatic model that we can support whatever aspects of existing systems and cloud systems that people need to do their business.”
The announcements are too numerous to sum up in a single article, but here are some of the highlights:
HP Converged Cloud
HP is helping customers to build hybrid cloud environments with a range of new software and management products. In particular, HP CloudSystem provides an integrated and open platform for enterprises to build and manage services across private, managed and public cloud environments.
The company has extended the “cloud bursting” capabilities of HP CloudSystem to deliver instant access to additional capacity through HP Cloud Services, Amazon Web Services and Savvis.
It is also now offering the software that runs on HP CloudSystem as a stand-alone product. This effectively means that customers can turn their in-house data centres into private clouds by running HP CloudSystem Matrix software on their existing infrastructure.
“Not only does it not lock you in, in terms of hardware, but it doesn't restrict you in terms of the operating systems, or in terms of the hypervisors,” said Greenway.
“IT can support a complex heterogeneous environment and allow you to manage that and allocate that using the tools that were previously just available as part of CloudSystem.”
A new version of HP Cloud Service Automation also offers a self-service portal to help customers increase agility and reduce costs when brokering and managing application and infrastructure services on HP CloudSystem-based solutions or third-party environments.
HP StoreOnce Backup
HP's new StoreOnce solutions can deliver backup performance of up to 100TB per hour and data recovery of up to 40TB per hour – which is up to three times and five times faster, respectively, than the closest competitive offering, according to HP.
The improved performance and recovery time means reduced capacity needs and bandwidth costs. StoreOnce is also integrated with HP Data Protector 7 Software, powered by Autonomy’s Intelligent Data Operating Layer (IDOL).
The company said that the new release allows customers to protect, find and recover information based on the meaning and concepts contained within the data.
“We not only backup and recover structured stuff, like you would expect in databases and application files, but all the unstructured stuff that Autonomy is so good at looking at – audio, video, email, social media files,” said David Chalmers, HP's CTO for Enterprise Servers Storage and Networking.
“They are all part of a standard backup and recovery, because we think that is crucial to how you're going to run large instances in the future.”
Chalmers said that, in a recent survey of HP customers, 40% were actively looking at what they can do or should do for new generations of backup and restore technology.
“This is the technical side of having far more data than they are used to; they need a new way of backing up and securing it, the old stuff just doesn't cut it any more.”
Application Performance Management
HP is expanding its capabilities with regards to DevOps, with the aim of breaking down IT silos and enabling continuous improvement and continuous rollout of new services onto the cloud.
Application Performance Management 9.2 for Converged Cloud helps helps customers manage across converged cloud environments, improving the performance management of cloud and mobile applications.
Meanwhile, the updated versions of HP Application Lifecycle Management (ALM) and HP Performance Center (PC) enable the continuous delivery of applications, with new Lab Management Automation and Continuous Application Performance Delivery features.
“Within enterprises, shifting towards the DevOps model is very much a reflection of the change from the monolithic application development to what Geoffrey Moore refers to as the 'systems of engagement' model, where more focused applications that are designed to deliver end-user needs,” said Greenway.
“People used to change their staff to match the way the computer works. Now the compute infrastructure supports the way the business needs to do its work.”
HP said that aligning an organisation's development environment more closely with its operational environment allows it to quickly deliver high-quality, innovative applications for hybrid and cloud environment.
Big Data in the cloud
HP subsidiary Autonomy has announced a series of cloud solutions based on HP’s Converged Cloud and Autonomy IDOL 10, including new capabilities for processing Hadoop data.
While Hadoop is a widely adopted and proven distributed file system, it does not understand the meaning of concepts contained in the information. By embedding the IDOL 10 engine in each Hadoop node, customers can make use of over 500 IDOL functions, including automatic categorisation, clustering, education, and hyperlinking.
Organisations can also combine Hadoop data with other enterprise and external data, using IDOL’s 400 connectors.
“We’ve done this in the past for Microsoft apps, SAP HANA, that sort of technology. Now we’re going to do it for Hadoop clusters as well,” said Chalmers.
“Pre-determined, pre-managed, pre-integrated hardware and software all comes in the box, making it much easier for the user to deploy – we’re taking about 30 minutes to get an 800 node cluster up and running.”
HP plans to publish a benchmark for Hadoop running on ProLiant Gen8 servers in next couple of weeks, which the company believes will show its leadership in Big Data processing performance.