Search for: All records

Unarbitrated contention over shared resources at different levels of the memory hierarchy represents a major source of temporal interference. Hardware manufacturers are increasingly more receptive to issues with temporal interference and are starting to propose concrete solutions to mitigate the problem. Intel Resource Director Technology (RDT) represents one such attempt. Given the wide adoption of Intel platforms, RDT features can be an invaluable asset for the consolidation of real-time systems on complex multi- and many-core machines. Unfortunately, to date, a systematic analysis of the capabilities introduced by the RDT framework has not yet been conducted. Moreover, no clear understanding has been matured about the implementation-specific behavior of RDT primitives across processor generations. And ultimately, the ability of RDT to provide real-time guarantees is yet to be established. In our work, we aim at conducting a systematic investigation of the RDT mechanisms from a real-time perspective. We experimentally evaluate the functionality and interpretability of RDT-aided allocation and monitoring controls across the two most recent processor generations. Our evaluations show that while some features like Cache Allocation Technology (CAT) yield promising results, the implementation of other primitives such as Memory Bandwidth Allocation (MBA) has much room for improvement. Moreover, in some cases, the presented interfaces range from blurry to incomplete, as is the case for MBA and Memory Bandwidth Monitoring (MBM). 

With the increasing dominance of SSDs for local storage, today's network mounted virtual disks can no longer offer competitive performance. We propose a Log-Structured Virtual Disk (LSVD) that couples log-structured approaches at both the cache and storage layer to provide a virtual disk on top of S3-like storage. Both cache and backend store are order-preserving, enabling LSVD to provide strong consistency guarantees in case of failure. Our prototype demonstrates that the approach preserves all the advantages of virtual disks, while offering dramatic performance improvements over not only commonly used virtual disks, but the same disks combined with inconsistent (i.e. unsafe) local caching. 

With the increasing dominance of SSDs for local storage, today's network mounted virtual disks can no longer offer competitive performance. We propose a Log-Structured Virtual Disk (LSVD) that couples log-structured approaches at both the cache and storage layer to provide a virtual disk on top of S3-like storage. Both cache and backend store are order-preserving, enabling LSVD to provide strong consistency guarantees in case of failure. Our prototype demonstrates that the approach preserves all the advantages of virtual disks, while offering dramatic performance improvements over not only commonly used virtual disks, but the same disks combined with inconsistent (i.e. unsafe) local caching. 

Kariz is a new architecture for caching data from datalakes accessed, potentially concurrently, by multiple analytic platforms. It integrates rich information from analytics platforms with global knowledge about demand and resource availability to enable sophisticated cache management and prefetching strategies that, for example, combine historical run time information with job dependency graphs (DAGs), information about the cache state and sharing across compute clusters. Our prototype supports multiple analytic frameworks (Pig/Hadoop and Spark), and we show that the required changes are modest. We have implemented three algorithms in Kariz for optimizing the caching of individual queries (one from the literature, and two novel to our platform) and three policies for optimizing across queries from, potentially, multiple different clusters. With an algorithm that fully exploits the rich information available from Kariz, we demonstrate major speedups (as much as 3×) for TPC-H and TPC-DS. 

Diagnosing performance problems in distributed applications is extremely challenging. A significant reason is that it is hard to know where to place instrumentation a priori to help diagnose problems that may occur in the future. We present the vision of an automated instrumentation framework, Pythia, that runs alongside deployed distributed applications. In response to a newly-observed performance problem, Pythia searches the space of possible instrumentation choices to enable the instrumentation needed to help diagnose it. Our vision for Pythia builds on workflow-centric tracing, which records the order and timing of how requests are processed within and among a distributed application's nodes (i.e., records their workflows). It uses the key insight that localizing the sources high performance variation within the workflows of requests that are expected to perform similarly gives insight into where additional instrumentation is needed.