hpe blog, uk & ireland
cancel
Showing results for 
Search instead for 
Did you mean: 

Finding the best infrastructure for your workload – one size does not fit all

Infrastructure screenshot.JPG

Finding the best infrastructure for your workload – one size does not fit all

Back in 2006 when HPE first introduced HPE BladeSystem, rack servers were all the rage, but the market still consumed tens of thousands of tower servers too. Organisations were getting to grips with virtualisation and starting the journey from physical servers to virtual servers – solving one problem of server utilisation but creating another with the largely uncontrolled growth of virtual machines.

The efficiencies of virtualisation

The HPE blade launch coincided with this rise of virtualisation, as organisations began to realise the scale of the challenge before them and the efficiencies they could make by reducing their sprawl of hardware devices. Back then, blades not only saved the day in terms of physical convergence and the associated savings from power and cooling, but allowed a contained environment of virtual machines, hosted on a physically close group of servers within a consolidated power footprint.

At that time all manner of shapes and sizes of form factors existed as vendors looked to either provide dense boxes to conserve rack space or large boxes to cram as much storage and network IO in as possible.  Many customers were still running 1Gb links between their servers and the idea of 10Gb connections across their network was merely an aspiration.

Jump forward to today and 10Gb is very much the norm with many organisations pushing 50Gb+ as they look to keep pace with the demands of applications while keeping an eye on network developments as 200Gb is on the horizon.  Not to mention the ultimate IO performance of Photonics which is promised to arrive in the next few years. 

The form factors of 14 years ago are still here, even if the popularity mix has changed.  Towers are down, rack servers may rule by volume, but blades most certainly still have their place. At HPE we look at workloads when designing infrastructure, to support the wide range of requirements that organisations have.

Does your data centre take advantage of today’s tech to have something for everybody?

Different workloads need very different support and as developments in tech evolve, we’re seeing just how efficient infrastructures can be when they are chosen for and tailored to the requirements of the workloads they are designed to support.

HPE SimpliVity for simple hyperconverged

In 2006 hyperconverged wasn’t yet commonplace, yet now it is the current buzzword in the industry.  The efficiencies of modern hardware and clever software mean that many HPE BladeSystem customers that are running virtualisation or VDI across one or two chassis over five years old will very likely find HPE SimpliVity – our hyperconverged range – a good fit as they plan to refresh their infrastructure and look for a cloud ready platform.

HPE SimpliVity is an intelligent yet simple hyperconverged platform that streamlines your IT systems, speeding applications and improving efficiency. It combines all your IT infrastructure and data services into a single solution, offering resilience with data protection, backing up and restoring in seconds. This single, integrated solution operates at a fraction of the cost of traditional systems and helps boost your users’ productivity.

HPE Apollo for high performance computing

In recent years we’ve seen a growth in demand for high density GPUs for HPC workloads that utilise specialised networking interconnects such as Infiniband. The HPE Apollo range of servers provides the ideal alternative for these very specific types of workloads. Optimised for high performance computing, these servers have a highly dense GPU capability or huge capacity of disk storage for Big Data, HPC or AI applications. This offers a variety of approaches for any customer looking to take a step up in performance from workloads they have creaking away on elderly blades.

Do you have tools that make your infrastructure easily accessible to developers in your organisation?

We recognise that applications are king, demanding a certain type of infrastructure to maximise their efficacy. Platforms change, and we need to keep developing solutions to provide the best performance for these ever-evolving applications in the most efficient way.

Just look at vintage cars which were great in their day. The modern versions are designed to go faster, be more resilient, have better control and management and you select the car most suitable for the surface you’re driving on. We can align the same practice to supporting application development and workloads; the infrastructures that were appropriate to support them 14 years ago were great building blocks to refine, develop and improve the systems we see today.

We are constantly working on new solutions, developing infrastructures to support the pace of change. With different products for evolving workloads, organisations have a much wider choice to select the solution that works for them.

 

Julian Keetch, Hybrid IT Business Development Manager

 

Get in touch

If you would like to pick our brains or discuss anything with our specialists, please get in touch today.

 

Previous blogs in this series:

Why you should get onboard with software-defined technology and HPE Synergy.

Catching up with HPE BladeSystem – The past, present and future of blades

0 Kudos
About the Author

Julian_Keetch

Events
Online Expert Days - 2020
Visit this forum and get the schedules for online Expert Days where you can talk to HPE product experts, R&D and support team members and get answers...
Read more
HPE at 2020 Technology Events
Learn about the technology events where Hewlett Packard Enterprise will have a presence in 2020
Read more
View all