Karl Taht's Research Paper Blog

Authors: Grant Ayers et al. Venue: ISCA 2019 Previous works have highlighted a significant frontend bottleneck is warehouse-scale computers (WSC). A variety of solutions have been proposed, both on real hardware and architectural papers, to mitigate the issue. This paper performs deep analysis across 90%+ of Google's entire fleet to perform a fine-grain analysis of when, where, and how frontend bottlenecks occur. AsmDB comprises of post-processed last branch record (LBR) data to form control flow probabilities and precise information about what instructions triggered I-cache misses. The analysis shows several core reasons for I-cache misses: large jump distances (either via function call or indirect branch) and cold code being brought into the cache (either via cache blocks or prefetching). The remainder of the paper transitions to being much more compiler focused. It describes a software prefetching algorithm which utilizes AsmDB information to inject prefetches at ideal poin...

Authors: Akshitha Sriraman, Abhishek Dhanotia, Thomas F. Wenisch Venue: ISCA 2019 This work comprises of two main parts: a detailed analysis and tool to improve course-grain parameters based on general application (microservice) behavior. The author's analyze workloads in Facebook's datacenter in the categories of Web, Feed, Ads, and Cache which have varying throughput and latency requirements. The data center workloads exhibit significant front-end stalls (instruction fetch misses), significant branch  mispredictions, and significant back-end stalls (mostly data cache misses). uSKU is presented as a tool which automates the process of parameter tuning in an effort to improve system optimization for specific classes of microservices. Core frequency, uncore frequency, core count, code-and-data prioritization, prefetchers, transparent and static huge pages are explored. Knobs are tested independently and thus to do not consider dependent effects (Gaussian process search seem...

Authors: Hongzi Mao, Malte Schwarzkopf, Shaileshh Bojja Venkatakrishnan, Zili Meng, Mohammad Alizadeh Venue: Proceedings of the ACM Special Interest Group on Data Communication This paper utilizes reinforcement learning to schedule learn a scheduling policy for Spark jobs. The scheduler has two main decisions: (i) what stage to schedule and (ii) how much parallelism to exploit for that stage. The RL problem is formulated as given the state of the cluster and DAG input, output a scheduling action. Reward is defined as -T x J where T is the time step and J is the number of jobs in the system. The decision making process is particularly difficult because an job can present a DAG of any shape for dependencies, yet, the neural network input is of fixed size. To solve this, a method based on graph convolutional neural networks [1] is used. The RL policy network predicts a composite action of stage of maximum parallelism level. To train the network in the case of continuous job arrivals, ...

Authors: Jayaraman Thiagarajan, Rushil Anirdh, Bhavya Kaikhura, Nikhil Jain, Tanzima Islam, Abhinav Bhatele, Jae-Seung Yeom, Todd Gamblin Venue: I EEE International Parallel and Distributed Processing Symposium (IPDPS) In the scope of HPC, machine learning is gaining increased traction to add in performance analysis and tuning. However, this approach includes a pipeline of data collection, data pre-processing, various machine learning algorithm testing, tuning, and then finally trying to understand the model. The paper states that while this process is repetitive, rarely can insights be reused from one domain to another. To address this void, the propose PADDLE. PADDLE has three key steps: deep feature extraction, model design, and visualization. The first step allows users to throw extensive amounts of data at the problem, and an automated solutions determines the key inputs, mapping them to a new feature space. The next step in paddle automatically tests a number of machine lea...

Authors: Benjamin Hindman, Andy Konwinski, Matei Zaharia, Ali Ghodsi, Anthony D. Joseph, Randy Katz, Scott Shenker, Ion Stoica Venue: NSDI 2011 Mesos is a thin management layer that allows various cluster computing frameworks to efficiently share resources. The two key principles leveraged by Mesos are its fine-grained resource sharing model at the level of tasks, and a distributed (decentralized) scheduling mechanism. The result is a framework which offers better overall system utilization, scalability to at least 50,000 nodes, and flexibility to port to many different (and future) frameworks. Fine-grain resource sharing is done via a notion of resource-offers, which are each a list of free resources on multiple slaves. The master decides how many resources to offer each framework, this distribution is done via a pluggable allocation module. A scheduler  registers with the master to be offered resources, and an executor process is launched on slave notes to run the framework'...

Authors: David Lo, Liqun Cheng, Rama Govindaraju, Parthasarathy Ranganathan, Christos Kozyrakis Venue:    ISCA 2015 This work presents a resource controller and scheduler that works to improve the throughput of best-effort tasks while preserving the SLO for latency-sensitive applications. The work combines tuning by addressing many fronts: core isolation (taskset), LLC isolation (CAT), power isolation (DVFS), and network traffic isolation (qdisc). They show that because of the unique property that these form a convex function, each can be optimized individually by understanding the current load of the system and available slack, which is polled by the top-level controller every 15 seconds. Overall, they increase machine utilization to 90% without sacrificing SLO agreements, which is defined for 60-second windows. The authors demonstrate three latency-critical workloads: websearch, ml_cluster, and memkeyval, which each stress different combinations of cache, bandwidth, pow...