Managing SMT Resource Usage through Speculative Instruction Window Weighting

Simultaneous multithreading processors dynamically share processor resources between multiple threads. In general, shared SMT resources may be managed explicitly, for instance, by dynamically setting queue occupation bounds for each thread as in the DCRA and Hill-Climbing policies. Alternatively, resources may be managed implicitly; that is, resource usage is controlled by placing the desired instruction mix in the resources. In this case, the main resource management tool is the instruction fetch policy which must predict the behavior of each thread (branch mispredictions, long-latency loads, etc.) as it fetches instructions.

Fetch Gating Control through Speculative Instruction Window Weighting

In a dynamic reordering superscalar processor, the front-end fetches instructions and places them in the issue queue. Instructions are then issued by the back-end execution core. Till recently, the front-end was designed to maximize performance without considering energy consumption. The front-end fetches instructions as fast as it can until it is stalled by a filled issue queue or some other blocking structure. This approach wastes energy: (i) speculative execution causes many wrong-path instructions to be fetched and executed, and (ii) back-end execution rate is usually less than its peak rate, but front-end structures are dimensioned to sustained peak performance. Dynamically reducing the front-end instruction rate and the active size of front-end structure (e.g. issue queue) is a required performance-energy trade-off. Techniques proposed in the literature attack only one of these effects.
In previous work, we have proposed Speculative Instruction Window Weighting (SIWW) [21], a fetch gating technique that allows to address both fetch gating and instruction issue queue dynamic sizing. SIWW computes a global weight on the set of inflight instructions. This weight depends on the number and types of inflight instructions (non-branches, high confidence or low confidence branches, ...). The front-end instruction rate can be continuously adapted based on this weight. This paper extends the analysis of SIWW performed in previous work. It shows that SIWW performs better than previously proposed fetch gating techniques and that SIWW allows to dynamically adapt the size of the active instruction queue.

The executive committee of the Greek Professional Association of Social Work in an age of austerity: examining its response

This paper explores the response by the Greek Association of Social Workers (SKLE) to Greece's current economic crisis. Socioeconomic conditions in Greece have deteriorated rapidly since the imposition of a Structural Adjustment Programme as a condition of the loan Troika provided to Greece to address its class-based public debt crisis. Interviews were conducted with SKLE Executive Committee members to examine SKLE's response in the context of newly raised inequalities. Research results show that SKLE recognised the negative consequences to both service users and its members. However, SKLE continues to reformulate its strategy mostly as a social partner. SKLE's previous strategy entailed amongst other things the analysis of policy proposals and participation in welfare related government committees. This strategy is no longer relevant because decision-making powers have been transferred to transnational bodies. This paper elaborates on these findings and discusses the barriers that prohibit SKLE from differentiation of its strategy. Although the research is country specific, it has implications for the broader global debate because professional associations must reformulate their strategies for better serving of both their constituents and the collective good based on the social justice mandate of the profession.

EU Social and Labour Rights and EU Internal Market Law: Study for the EMPL Committee

EU Social and Labour Rights have developed incrementally, originally through a set of legislative initiatives creating selective employment rights, followed by a non-binding Charter of Social Rights. Only in 2009, social and labour rights became legally binding through the Charter of Fundamental Rights for the European Union (CFREU). By contrast, the EU Internal Market - an area without frontiers where goods, persons, services and capital can circulate freely – has been enshrined in legally enforceable Treaty provisions from 1958. These comprise the economic freedoms guaranteeing said free circulation and a system ensuring that competition is not distorted within the Internal Market (Protocol 27 to the Treaty of Lisbon). Tensions between Internal Market law and social and labour rights have been observed in analyses of EU case law and legislation. This study explores responses by socio-economic and political actors at national and EU levels to such tensions, focusing on collective labour rights, rights to fair working conditions and rights to social security and social assistance (Articles 12, 28, 31, 34 Charter of Fundamental Rights for the European Union). On the basis of the current Treaties and the CFREU, the constitutionally conditioned Internal Market emerges as a way to overcome the perception that social and labour rights limit Internal Market law, or vice versa. On this basis, alternative responses to perceived tensions are proposed, focused on posting of workers, furthering fair employment conditions through public procurement and enabling effective collective bargaining and industrial action in the Internal Market.

Methods for Optimal Work Instruction Delivery: Pictograms v Images

This paper focuses on quantifying the benefits of pictogram based instructions relative to static images for work instruction delivery. The assembly of a stiffened aircraft panel has been used as an exemplar for the work which seeks to address the challenge of identifying an instructional mode that can be location or language neutral while at the same time optimising assembly build times and maintaining build quality. Key performance parameters measured using a series of panel build experiments conducted by two separate groups were: overall build time, the number of subject references to instructional media, the number of build errors and the time taken to correct any mistakes. Overall build time for five builds for a group using pictogram instructions was about 20% lower than for the group using image based instructions. Also, the pictogram group made fewer errors. Although previous work identified that animated instructions result in optimal build times, the language neutrality of pictograms as well as the fact that they can be used without visualisation hardware mean that, on balance, they have broader applicability in terms of transferring assembly knowledge to the manufacturing environment.