HARNESS integrates heterogeneous hardware and network technologies into data centre platforms, vastly increasing performance, reducing energy consumption, and lowering cost profiles for important and high-value cloud applications such as real-time business analytics and the geosciences.
Homogeneous, commodity cloud computing has reached its limits
The dominant approach in offering cloud services today is based on homogeneous commodity resources: large numbers of inexpensive machines, interconnected by off-the-shelf networking equipment, supported by stock disk drives. However, cloud service providers are unable to use this platform to satisfy the requirements of many important and high-value classes of applications.
Today’s cloud platforms are missing out on the revolution in new hardware and network technologies for realising vastly richer computational, communication, and storage resources. Technologies such as Field Programmable Gate Arrays (FPGA), General-Purpose Graphics Processing Units (GPGPU), programmable network routers, and solid-state disks promise increased performance, reduced energy consumption, and lower cost profiles. However, their heterogeneity and complexity makes integrating them into the standard Platform as a Service (PaaS) framework a fundamental challenge.
The HARNESS project brings innovative and heterogeneous resources into cloud platforms through a rich programme of research, validated by commercial and open source case studies.
The HARNESS vision
The HARNESS project advances the state of the art in cloud data centre design so that:
- cloud providers can profitably offer and manage the tenancy of specialised hardware and network technologies much as they do with today’s commodity resources, and
- software engineers can seamlessly, flexibly, and cost-effectively integrate specialised hardware and network technologies into the design and execution of their cloud- hosted applications.
HARNESS develops an enhanced PaaS software stack that brings new degrees of freedom to cloud resource allocation and optimisation. A cloud application is seen to consist of components, some of which have multiple implementations. Applications express their computing needs to the cloud platform, as well as the price they are prepared to pay for various levels of service. This expression of needs and constraints builds upon what can be expressed through today’s simple counts of virtual machines or amounts of storage, to encompass the specific and varied new factors characteristic of specialised hardware and network technologies. The cloud platform will have access to a variety of resources to which it can map the components. A flexible application may potentially be deployed in many different ways over these resources, each option having its own cost, performance, and usage characteristics.
Specialised technologies are virtualised into resources that can be managed and accessed at the platform level. The idea is to provide flexibility to the platform as to which, when, and how many resources are used, and to separate that concern from the low-level deployment and monitoring of the concrete technology elements. Associated with the virtualised resources are policies that govern how allocation and optimisation decisions are made. Also associated with these resources are facilities to track their capacity, usage, and general availability.
The public cloud services provider market is projected to reach nearly $22 billion by 2015. HARNESS will enable those providers to offer new levels of service to cloud applications at the same time as it opens a new market to the purveyors of specialised hardware and network technologies.