A5034439600- Cloud Computing and Resource Management
- Parallel Computing and Optimization Techniques
- Distributed and Parallel Computing Systems
- IoT and Edge/Fog Computing
- Advanced Data Storage Technologies
- Caching and Content Delivery
- Non-Destructive Testing Techniques
- Machine Learning and ELM
- Graph Theory and Algorithms
- Advanced Neural Network Applications
- Distributed systems and fault tolerance
- Muscle Physiology and Disorders
- Software System Performance and Reliability
- Neurogenetic and Muscular Disorders Research
- Cardiomyopathy and Myosin Studies
Universitat Politècnica de València
2018-2024
Universitat Politècnica de Catalunya
2023
Max Delbrück Center
2008
Charité - Universitätsmedizin Berlin
2008
University Health Network
2008
Ospedale Regionale di Mendrisio
2008
Deutsches Herzzentrum München
2008
Left ventricular noncompaction constitutes a primary cardiomyopathy characterized by severely thickened, 2-layered myocardium, numerous prominent trabeculations, and deep intertrabecular recesses. The genetic basis of this is still largely unresolved. We speculated that mutations in sarcomere protein genes known to cause hypertrophic dilated may be associated with left noncompaction.Mutational analysis cohort 63 unrelated adult probands no other congenital heart anomalies was performed...
The increasing popularity of cloud computing has forced providers to build economies scale meet the growing demand. Nowadays, data-centers include thousands physical machines, each hosting many virtual machines (VMs), which share main system resources, causing interference that can significantly impact on performance. Frequently, these run latency-critical workloads, whose performance is determined by tail latency, very sensitive co-running workloads. To prevent QoS violations, adopt...
The Last Level Cache (LLC) plays a key role in the system performance of current multi-cores by reducing number long latency main memory accesses. inter-application interference at this shared resource, however, can lead to undesired situations regarding and fairness. Recent approaches have successfully addressed fairness turnaround time (TT) commercial processors. Nevertheless, these must face sustaining performance, which is challenging. This work makes two contributions. LLC behaviors...
Multithreaded latency-critical applications represent an important subset of workloads running on public cloud systems. Most these systems deploy powerful computing servers including Intel Hyper-Threading processors. Understanding how performance is affected by the consumption main system resources a major concern for providers in order to devise virtualization strategies that improve efficiency. With this aim, paper first characterizes impact QPS tail latency, analyzing different scenarios...
Cloud systems deploy a wide variety of shared resources and host large number tenant applications. To perform cloud research, small experimental platform is commonly used, which hides the huge system complexity provides flexibility. Despite being simpler, this should include main components (hardware software) to provide representative results. A set platforms have spread in recent years; however, most them only major component or lack deployment virtual machines (VMs) isolation. This paper...
Graph processing is increasingly adopted to solve problems that span many application domains, including scientific computing, social networks, and big-data analytics. These applications present particular features (huge working sets irregular scalability) make the default Linux scheduler, which adopts a time-sharing policy provide fair perform poorly when co-locating multiple graph in same processor. This work focuses on maximizing processor utilization, major concern of current data...
The microarchitecture of general-purpose processors is continuously evolving to adapt the new computation and memory demands incoming workloads. In this regard, circuitry added execute specific instructions like vector multiplication or string operations. These enhancements support multiple threads per core make simultaneous multithreading (SMT) dominate market for data center processors. Regarding emerging workloads, machine learning taking an important role in many research domains...
Current server processors have redistributed the cache hierarchy space over previous generations. The private L2 has been made larger and shared last level caches (LLC) smaller but designed as non-inclusive to reduce number of replicated blocks. As a result, new organization shrinks per-core area.
Simultaneous multithreading processors are dominating the High Computing Performance market. Among these processors, those supporting only two threads being most widely deployed in current systems, thus, compete at run-time for intra-core resources. The performance of can be boosted by selecting symbiotic applications to executed on same core, which reduces inter-application interference considerably. In this letter we propose Hy-Sched, an scheduling algorithm that exploits symbiosis make...
Abstract Research on resource management focuses optimizing system performance and energy efficiency by distributing shared resources like processor cores, caches, main memory among competing applications. This research spans a wide range of applications, including those from high-performance computing, machine learning, mobile computing. Existing frameworks often simplify concentrating specific characteristics, such as the architecture computing nodes, monitoring, representative workloads....
Understanding inter-VM interference is of paramount importance to provide a sound knowledge and understand where performance degradation comes from in the current public cloud. With this aim, paper devises workload taxonomy that classifies applications according how major system resources affect their (e.g., tail latency) as function level load QPS). After that, we present three main studies addressing concerns improve cloud performance: impact on performance, hyper-threading limiting last...