Automatic Data Allocation and Buffer Management for Multi-GPU Machines

Multi-GPU machines are being increasingly used in high-performance computing. Each GPU in such a machine has its own memory and does not share the address space either with the host CPU or other GPUs. Hence, applications utilizing multiple GPUs have to manually allocate and manage data on each GPU. Existing works that propose to automate data allocations for GPUs have limitations and inefficiencies in terms of allocation sizes, exploiting reuse, transfer costs, and scalability. We propose a scalable and fully automatic data allocation and buffer management scheme for affine loop nests on multi-GPU machines. We call it the Bounding-Box-based Memory Manager (BBMM). BBMM can perform at runtime, during standard set operations like union, intersection, and difference, finding subset and superset relations on hyperrectangular regions of array data (bounding boxes). It uses these operations along with some compiler assistance to identify, allocate, and manage data required by applications in terms of disjoint bounding boxes. This allows it to (1) allocate exactly or nearly as much data as is required by computations running on each GPU, (2) efficiently track buffer allocations and hence maximize data reuse across tiles and minimize data transfer overhead, and (3) and as a result, maximize utilization of the combined memory on multi-GPU machines. BBMM can work with any choice of parallelizing transformations, computation placement, and scheduling schemes, whether static or dynamic. Experiments run on a four-GPU machine with various scientific programs showed that BBMM reduces data allocations on each GPU by up to 75% compared to current allocation schemes, yields performance of at least 88% of manually written code, and allows excellent weak scaling.

Dasher - A data entry interface using continuous gestures and language models

Existing devices for communicating information to computers are bulky, slow to use, or unreliable. Dasher is a new interface incorporating language modelling and driven by continuous two-dimensional gestures, e.g. a mouse, touchscreen, or eye-tracker. Tests have shown that this device can be used to enter text at a rate of up to 34 words per minute, compared with typical ten-finger keyboard typing of 40-60 words per minute. Although the interface is slower than a conventional keyboard, it is small and simple, and could be used on personal data assistants and by motion-impaired computer users.

Summary of time requirements and potential improvements for Marine Recorder data entry.

The contract work has demonstrated that older data can be assessed and entered into the MR format. Older data has associated problems but is retrievable. The contract successfully imported all datasets as required. MNCR survey sheets fit well into the MR format. The data validation and verification process can be improved. A number of computerised short cuts can be suggested and the process made more intuitive. Such a move is vital if MR is to be adopted as a standard by the recording community both on a voluntary level and potentially by consultancies.

Adaptive tool for automatic data collection of real electricity markets

The study of electricity markets operation has been gaining an increasing importance in last years, as result of the new challenges that the electricity markets restructuring produced. This restructuring increased the competitiveness of the market, but with it its complexity. The growing complexity and unpredictability of the market’s evolution consequently increases the decision making difficulty. Therefore, the intervenient entities are forced to rethink their behaviour and market strategies. Currently, lots of information concerning electricity markets is available. These data, concerning innumerous regards of electricity markets operation, is accessible free of charge, and it is essential for understanding and suitably modelling electricity markets. This paper proposes a tool which is able to handle, store and dynamically update data. The development of the proposed tool is expected to be of great importance to improve the comprehension of electricity markets and the interactions among the involved entities.

Automatic (data-based) learning of expressive indicators for machine failure detection : a proposal

Máster Universitario en Sistemas Inteligentes y Aplicaciones Numéricas en Ingeniería (SIANI)

Gamification e data entry: un caso di studio

Il gioco è un concetto che accompagna la vita di innumerevoli specie animali in forme, modi e tempi differenti. L’uomo scopre il gioco sin dai primi mesi di vita. Con l’obiettivo di migliorare la condizione emotiva dell'uomo nello svolgimento delle azioni quotidiane, nasce negli ultimi anni la gamification. Il termine consta nell’integrazione delle tecniche di progettazione dei giochi in contesti esterni ai giochi. Consiste nel progettare ponendo particolare attenzione sul coinvolgimento dell’utente per renderlo capace di sperimentare le emozioni tipiche dello svago: fierezza per le proprie azioni, qualunque esse siano. Gli ambiti di applicazione sono innumerevoli. Questa tesi si concentra sullo studio del contesto aziendale, focalizzandosi sulle mansioni di data entry, allo scopo di creare una piattaforma completa, composta da strumenti informatici ed elementi di gioco, che possa aumentare il coinvolgimento dei dipendenti nel proprio lavoro. Si è scelto questo tipo di attività in quanto composta da incarichi facilmente misurabili e allo stesso tempo poco appassionanti per il dipendente perché altamente meccanici e ripetitivi. La sperimentazione in questo ambito permette quindi di valutare con certezza matematica se i miglioramenti introdotti dall'integrazione delle tecniche di gamification nello stato d’animo dei dipendenti hanno anche la conseguenza di aumentare la produttività, verificando quindi se una piattaforma ludicizzata possa essere auto-sostenibile in ambito aziendale. Al termine della tesi si giungerà ad ottenere il progetto di un sistema completo, composto da software ed attività extra-informatiche, che i dipendenti valuteranno con un questionario. La piattaforma otterrà buoni voti necessitando principalmente di un maggior apporto contenutistico e del contributo professionale di un esperto progettista di giochi perché abbia le potenzialità per diventare un caso di successo.

Automatic Data Normalization and Parameterization for Optical Motion Tracking

Methods for optical motion capture often require timeconsuming manual processing before the data can be used for subsequent tasks such as retargeting or character animation. These processing steps restrict the applicability of motion capturing especially for dynamic VR-environments with real time requirements. To solve these problems, we present two additional, fast and automatic processing stages based on our motion capture pipeline presented in [HSK05]. A normalization step aligns the recorded coordinate systems with the skeleton structure to yield a common and intuitive data basis across different recording sessions. A second step computes a parameterization based on automatically extracted main movement axes to generate a compact motion description. Our method does not restrict the placement of marker bodies nor the recording setup, and only requires a short calibration phase.

Review of automatic data processing developments in the Federal Government.

Mode of access: Internet.