IMC2: Instrumentation Measurement and Control for the Cloud

The Internet landscape is changing rapidly, from a completely decentralised paradigm where distinct services were offered by different providers in a fully distributed and decentralised way, to a unified ICT environment where data, storage, and processing resources are co-located in the Cloud, and offered alongside connectivity. Although Cloud services and the underlying communication infrastructures are built on top of commodity Internet mechanisms (transport protocols, IP switching, multipath routing, etc.), it becomes apparent that the performance-agnostic and slow-converging operational assumptions of today's data communications are challenged by the new unified technological and business model. Massive overprovisioning of fully distributed resources that are managed in distinct and often long timescales (e.g., traffic aggregates over backbone networks) is not sustainable in an environment where connectivity and system resources need to be managed by a single unified ICT provider over a centralised infrastructure and in very short timescales. Cloud providers need to maximise return-on-investment from their infrastructures through rapid provisioning and elastic resource management, offering predictable services while operating at higher utilisation thresholds.

In order to achieve these goals, in this project we will design and develop an always-on Instrumentation, Measurement, and Control (IMC) framework that will dynamically and adaptively provision unified resources in a unified manner and in short timescales. Evidence has shown that distinct control loops typically employed to manage different resources in different timescales can themselves constitute factors of performance degradation over unified Cloud environments. For example, network-agnostic placement and migration of virtual machines can itself cause congestion in the underlying Data Centre topology. We will therefore revisit the one-dimensional, static or pseudo-random control loops that are typically employed over Cloud topologies, and develop an adaptive closed-loop system that will manage both server and network resources synergistically, in short timescales and based on temporal topology-wide performance. In doing so, we will exploit often controversial concepts such as non-shortest path routing for increasing load balancing while meeting flow completion deadlines, and network-aware dynamic virtual machine migration, to demonstrate the feasibility and also the benefits of combinatorial resource provisioning in achieving global performance optimisation and in increasing the usable capacity of future networks and services. One of the key aims of the proposed research is to investigate and to demonstrate the applicability of measurement-based processes to control and to admit resources in a unified manner and at appropriate, short timescales. Through the necessary system and network node instrumentation, we will devise a logically-centralised measurement and control closed-loop architecture that will be an integral part of the underlying infrastructure's data forwarding operation. The long-term impact of such endeavour will be to revisit the currently disjoint data and control planes in packet communications, and to transform next generation networked infrastructures from performance-agnostic to adaptive and self-managed, through synergy across the different layers and planes of the architecture.

The proposed research will be carried out at the University of Glasgow, and experiments will be conducted over a purpose-built programmable Cloud services testbed infrastructure, partly supported by EPSRC's first grant scheme and partly through a generous contribution from the host institution. The research will be conducted in close collaboration with Onyx Group, Microsoft Research and JANET(UK).

Besides the documented academic impact, this project has excellent potential for both direct and indirect commercial exploitation, partly evidenced by the accompanying letters of support and the close engagement with industry and R&D. Directly, the results of this work will have immediate and long-term applicability for the following beneficiaries:

Cloud service providers and Data Centre operators - Through the always-on combinatorial optimisation of server and network resource usage, results from this research will have an immediate impact in increasing the usable capacity of Cloud Data Centres, and therefore significantly improve return-on-investment for infrastructure and service providers. This will enable providers to absorb short-term increases and fluctuations in traffic demands without investing in excess capacity, or penalising performance and violating service level agreements. The immense importance of this is evident when one considers the drop in revenues that can be caused by even a marginal increase in service response time (e.g., Amazon's EC2 1% drop in sales resulting from 100 ms additional latency).
We will liaise closely with Onyx Group, a national Data Centre operator and service provider with significant regional presence in Glasgow who is supporting this work, and will provide us with operational data and actual topology/provisioning characteristics of private Cloud environments. This will ensure that results arising from this work will be relevant and directly applicable to the growing business community of Cloud service providers in the immediate 1-5 year horizon. With the increasing number of private Cloud companies, this research has excellent potential to have significant impact in the UK's economic competitiveness.
In addition, consultation with Microsoft Research will ensure that the research will also remain relevant and influence the future design of global, public Cloud providers that play a key role in shaping the global service provisioning trends in the long term.

Unified ICT and ISP providers - With bandwidth becoming a commodity and profit margins narrowing, traditional ISP providers are currently investing in unified ICT services on top of connectivity. Through direct liaising with JANET(UK), the UK's national research and education network provider, results from this work will enable ISPs to exploit programmable and self-managed service provisioning, while rolling out and expanding next generation unified ICT infrastructures for the next 5-10 years.

Equipment and software vendors - Results from this work will shed new light in the instrumentation capabilities of network and server architectures with always-on measurement and control functionality. Software prototypes resulting from this project will be of immediate benefit to equipment and software vendors, especially the virtualisation, OpenFlow, and programmable switch manufacturer industries (e.g., NetFPGA, Solarflare).

End users and corporate customers - The proposed research will result in improving the performance, dependability and predictability of outsourced ICT, benefiting end-users but also corporate customers of Cloud services, such as the private sector and the government. Facilitating the cost-effective and wide-spread adoption of such services will have an immediate and long-term indirect societal and economic impact.

The RA working on this project will develop research and development skills in cutting-edge networking and system technologies, while working on a timely research topic with growing international popularity.
Indirectly, this research will have a long-term impact on the design of future unified distributed systems architectures that will integrate always-on measurement and control functionality, enabling truly extensible, dynamically adaptive and self-managed services. This will lead to novel business models regarding the manufacturing, usage and charging for ICT services in the future.