Toward a Scalable Working Set Size Estimation Method and Its Application for Chip Multiprocessors

It is essential to accurately estimate the working set size (WSS) of an application for various optimizations such as to partition cache among virtual machines or reduce leakage power dissipated in an over-allocated cache by switching it OFF. However, the state-of-the-art heuristics such as average memory access latency (AMAL) or cache miss ratio (CMR) are poorly correlated to the WSS of an application due to 1) over-sized caches and 2) their dispersed nature. Past studies focus on estimating WSS of an application executing on a uniprocessor platform. Estimating the same for a chip multiprocessor (CMP) with a large dispersed cache is challenging due to the presence of concurrently executing threads/processes. Hence, we propose a scalable, highly accurate method to estimate WSS of an application. We call this method ``tagged WSS (TWSS)'' estimation method. We demonstrate the use of TWSS to switch-OFF the over-allocated cache ways in Static and Dynamic NonUniform Cache Architectures (SNUCA, DNUCA) on a tiled CMP. In our implementation of adaptable way SNUCA and DNUCA caches, decision of altering associativity is taken by each L2 controller. Hence, this approach scales better with the number of cores present on a CMP. It gives overall (geometric mean) 26% and 19% higher energy-delay product savings compared to AMAL and CMR heuristics on SNUCA, respectively.

Hybrid Working Set Algorithm for SVM Learning With a Kernel Coprocessor on FPGA

Support vector machines (SVM) are a popular class of supervised models in machine learning. The associated compute intensive learning algorithm limits their use in real-time applications. This paper presents a fully scalable architecture of a coprocessor, which can compute multiple rows of the kernel matrix in parallel. Further, we propose an extended variant of the popular decomposition technique, sequential minimal optimization, which we call hybrid working set (HWS) algorithm, to effectively utilize the benefits of cached kernel columns and the parallel computational power of the coprocessor. The coprocessor is implemented on Xilinx Virtex 7 field-programmable gate array-based VC707 board and achieves a speedup of upto 25x for kernel computation over single threaded computation on Intel Core i5. An application speedup of upto 15x over software implementation of LIBSVM and speedup of upto 23x over SVMLight is achieved using the HWS algorithm in unison with the coprocessor. The reduction in the number of iterations and sensitivity of the optimization time to variation in cache size using the HWS algorithm are also shown.

Modelling of the Subarctic North Pacific Circulation (Report of Working Group 7)

This review focuses on modelling ocean circulation and its variability in the subarctic North Pacific; it addresses issues specific to that region, and not the subject of ocean modelling in general. The performance of existing models is assessed in relation to observations in the upper ocean, intermediate waters and deep/abyssal waters. (PDF contains 87 pages)

Coastal Pelagic Fishes (Report of Working Group 3); Subarctic Gyre (Report of Working Group 6)

This is a report of PICES Working Group 3 (Coastal Pelagic Fishes) for 1993 and the first Annual Report of the Subarctic Gyre Working Group (WG-6). (PDF contains 131 pages)

Destruction of Chemical Agents in a Plasma Reactor with Working Gas of Hydrogen

Review of selected organisational IDs and development of use cases for the Jisc CASRAI-UK Organisational Identifiers Working Group

This project aimed to investigate and review candidates for providing an authoritative, widely used unique identifier for organisations involved in research in the UK. Specifically, we undertook to: ● clarify a representative but not comprehensive set of use cases for the UK research community to use organisational identifiers (orgIDs); ● survey and interview a small number of well-informed people in the field in order to create and prioritise a list of desirable features for the provision of orgIDs and potential services built around them; ● check the use cases and these required features against four possible candidate orgIDs and their providers; ● inform the Working Group of our conclusions and, if appropriate, make recommendations for adoption by the UK research community.

Ways of working smarter not harder

A look at best practice, good practice and administrative workflows

Telling stories, understanding lives, working toward change

Stories are helping us learn more about the livelihoods of the fishers and farmers with whom we work in eastern India. We are engaged with these communities in processes and activities aimed at improving their lives and promoting changes in government policy and service delivery in aquaculture and fisheries. Stories are told in several languages by women and men who fish and farm, about their lives, their livelihoods and significant changes they have experienced. We also record stories as narrated to us by colleague-informants. The written and spoken word, photographs, drawings and films – all are used to document the stories of people’s lives, sometimes prompted by questions as simple as “What do people talk about in the village?” Through the power of language, stories can be an entry point into livelihoods programming, monitoring and evaluation, conflict transformation and ultimately a way of giving life to a rights-based approach to development. (PDF contains 10 pages).

The Kaipara story: a closer look at the benefits of working together, the evolution of a federation of aquaculture self-help groups and a one-stop aqua shop in rural West Bengal

What is special about Kaipara is that most recently, they have founded a federation of Self-Help Groups that work together to develop their own support network and to draw in the support of others. This is a sophisticated ‘home-grown’ support infrastructure that is the subject of this story. (Pdf contains 8 pages).

Assessing the sustainability of brackish water aquaculture systems in the Phillipines. Working Paper 1: A typology of brackish-water pond aquaculture systems in the Philippines

Aquaculture in the Philippines is a long-standing activity but has witnessed relatively recent, rapid, technical change with the introduction of hatchery technology and commercial feed-mills changing the production possibilities for a fishpond operator. We are confronted with a diversity of aquaculture practices in the coastal areas of the Philippines, with new technologies being incorporated into more traditional systems. As a first step to understanding the sector, we therefore present a typology of farming systems with the motivation of generating domains (farm “types”) over which we can compare performance on a number of indicators. Our typology, restricted to brackish-water pond systems, is constructed using multivariate methods (principal components analysis, cluster analysis). Eight variables are used relating to the management of the farm across all the major factors of production. A stratified net sample of 136 observations provides the data for the analysis, from a farm-level survey carried out between January and June 2003 in the two main brackish-water production regions in the Philippines. We define five farm types from this analysis. In later work we will show how the use of this typology can be used for comparative study of economic, social and ecological performance at the farm-level. [PDF contains 42 pages]