arXiv subjects

Roy Friedman

Publications and source records attributed to Roy Friedman.

At least 37 records · Page 2Linked to original sources

arXiv2021

Limited Associativity Makes Concurrent Software Caches a Breeze

Software caches optimize the performance of diverse storage systems, databases and other software systems. Existing works on software caches automatically resort to fully associative cache designs. Our work shows that limited associativity caches are a promising direction for concurrent software caches. Specifically, we demonstrate that limited associativity enables simple yet efficient realizations of multiple cache management schemes that can be trivially parallelized. We show that the obtained hit ratio is usually similar to fully associative caches of the same management policy, but the throughput is improved by up to X5 compared to production-grade caching libraries, especially in multi-threaded executions.

cs.AR↗

arXiv2021

Lightweight Robust Size Aware Cache Management

Modern key-value stores, object stores, Internet proxy caches, as well as Content Delivery Networks (CDN) often manage objects of diverse sizes, e.g., blobs, video files of different lengths, images with varying resolution, and small documents. In such workloads, size-aware cache policies outperform size-oblivious algorithms. Unfortunately, existing size-aware algorithms tend to be overly complicated and computationally~expensive. Our work follows a more approachable pattern; we extend the prevalent (size-oblivious) TinyLFU cache admission policy to handle variable sized items. Implementing our approach inside two popular caching libraries only requires minor changes. We show that our algorithms yield competitive or better hit-ratios and byte hit-ratios compared to the state of the art size-aware algorithms such as AdaptSize, LHD, LRB, and GDSF. Further, a runtime comparison indicates that our implementation is faster by up to x3 compared to the best alternative, i.e., it imposes much lower CPU overhead.

cs.OS↗

arXiv2021

Posterior Sampling for Image Restoration using Explicit Patch Priors

Almost all existing methods for image restoration are based on optimizing the mean squared error (MSE), even though it is known that the best estimate in terms of MSE may yield a highly atypical image due to the fact that there are many plausible restorations for a given noisy image. In this paper, we show how to combine explicit priors on patches of natural images in order to sample from the posterior probability of a full image given a degraded image. We prove that our algorithm generates correct samples from the distribution $p(x|y) \propto \exp(-E(x|y))$ where $E(x|y)$ is the cost function minimized in previous patch-based approaches that compute a single restoration. Unlike previous approaches that computed a single restoration using MAP or MMSE, our method makes explicit the uncertainty in the restored images and guarantees that all patches in the restored images will be typical given the patch prior. Unlike previous approaches that used implicit priors on fixed-size images, our approach can be used with images of any size. Our experimental results show that posterior sampling using patch priors yields images of high perceptual quality and high PSNR on a range of challenging image restoration problems.

cs.CV↗

arXiv2021

Accelerating Big-Data Sorting Through Programmable Switches

Sorting is a fundamental and well studied problem that has been studied extensively. Sorting plays an important role in the area of databases, as many queries can be served much faster if the relations are first sorted. One of the most popular sorting algorithm in databases is merge sort. In modern data-centers, data is stored in storage servers, while processing takes place in compute servers. Hence, in order to compute queries on the data, it must travel through the network from the storage servers to the compute servers. This creates a potential for utilizing programmable switches to perform partial sorting in order to accelerate the sorting process at the server side. This is possible because, as mentioned above, data packets pass through the switch in any case on their way to the server. Alas, programmable switches offer a very restricted and non-intuitive programming model, which is why realizing this is not-trivial. We devised a novel partial sorting algorithm that fits the programming model and restrictions of programmable switches and can expedite merge sort at the server. We also utilize built-in parallelism in the switch to divide the data into sequential ranges. Thus, the server needs to sort each range separately and then concatenate them to one sorted stream. This way, the server needs to sort smaller sections and each of these sections is already partially sorted. Hence, the server does less work, and the access pattern becomes more virtual-memory friendly. We evaluated the performance improvements obtained when utilizing our partial sorting algorithm over several data stream compositions with various switch configurations. Our study exhibits an improvement of 20%-75% in the sorting run-time when using our approach compared to plain sorting on the original stream.

cs.DB↗

arXiv2020

Jiffy: A Fast, Memory Efficient, Wait-Free Multi-Producers Single-Consumer Queue

In applications such as sharded data processing systems, sharded in-memory key-value stores, data flow programming and load sharing applications, multiple concurrent data producers are feeding requests into the same data consumer. This can be naturally realized through concurrent queues, where each consumer pulls its tasks from its dedicated queue. For scalability, wait-free queues are often preferred over lock based structures. The vast majority of wait-free queue implementations, and even lock-free ones, support the multi-producer multi-consumer model. Yet, this comes at a premium, since implementing wait-free multi-producer multi-consumer queues requires utilizing complex helper data structures. The latter increases the memory consumption of such queues and limits their performance and scalability. Additionally, many such designs employ (hardware) cache unfriendly memory access patterns. In this work we study the implementation of wait-free multi-producer single-consumer queues. Specifically, we propose Jiffy, an efficient memory frugal novel wait-free multi-producer single-consumer queue and formally prove its correctness. We then compare the performance and memory requirements of Jiffy with other state of the art lock-free and wait-free queues. We show that indeed Jiffy can maintain good performance with up to 128 threads, delivers up to 50% better throughput than the next best construction we compared against, and consumes ~90% less memory.

cs.DB↗

arXiv2019

FireLedger: A High Throughput Blockchain Consensus Protocol

Blockchains are distributed secure ledgers to which transactions are issued continuously and each block of transactions is tightly coupled to its predecessors. Permissioned blockchains place special emphasis on transactions throughput. In this paper we present FireLedger, which leverages the iterative nature of blockchains in order to improve their throughput in optimistic execution scenarios. FireLedger trades latency for throughput in the sense that in FireLedger the last f + 1 blocks of each node's blockchain are considered tentative, i.e., they may be rescinded in case one of the last f + 1 blocks proposers was Byzantine. Yet, when optimistic assumptions are met, a new block is decided in each communication step, which consists of a proposer that sends only its proposal and all other participants are sending a single bit each. Our performance study demonstrates that in a single Amazon data-center, FireLedger running on 10 mid-range Amazon nodes obtains a throughput of up to 160K transactions per second for (typical Bitcoin size) 512 bytes transactions. In a 10 nodes Amazon geo-distributed setting with 512 bytes transactions, FireLedger obtains a throughput of 30K tps. Moreover, on higher end Amazon machines, FireLedger obtains $20%-600%$ better throughput than state of the art protocols like HotStuff and BFT-SMaRt, depending on the exact configuration.

cs.DC↗

arXiv2019

A Generic Efficient Biased Optimizer for Consensus Protocols

Consensus is one of the most fundamental distributed computing problems. In particular, it serves as a building block in many replication based fault-tolerant systems and in particular in multiple recent blockchain solutions. Depending on its exact variant and other environmental assumptions, solving consensus requires multiple communication rounds. Yet, there are known optimistic protocols that guarantee termination in a single communication round under favorable conditions. In this paper we present a generic optimizer than can turn any consensus protocol into an optimized protocol that terminates in a single communication round whenever all nodes start with the same predetermined value and no Byzantine failures occur (although node crashes are allowed). This is regardless of the network timing assumptions and additional oracle capabilities assumed by the base consensus protocol being optimized. In the case of benign failures, our optimizer works whenever the number of faulty nodes $f<n/2$. For Byzantine behavior, our optimizer's resiliency depends on the validity variant sought. In the case of classical validity, it can accommodate $f<n/4$ Byzantine failures. With the more recent external validity function assumption, it works whenever $f<n/3$. Either way, our optimizer only relies on oral messages, thereby imposing very light-weight crypto requirements.

cs.DC↗

arXiv2018

Heavy Hitters over Interval Queries

Heavy hitters and frequency measurements are fundamental in many networking applications such as load balancing, QoS, and network security. This paper considers a generalized sliding window model that supports frequency and heavy hitters queries over an interval given at \emph{query time}. This enables drill-down queries, in which the behavior of the network can be examined in finer and finer granularities. For this model, we asymptotically improve the space bounds of existing work, reduce the update and query time to a constant, and provide deterministic solutions. When evaluated over real Internet packet traces, our fastest algorithm processes packets $90$--$250$ times faster, serves queries at least $730$ times quicker and consumes at least $40\%$ less space than the known method.

cs.DS↗

arXiv2018

Give Me Some Slack: Efficient Network Measurements

Many networking applications require timely access to recent network measurements, which can be captured using a sliding window model. Maintaining such measurements is a challenging task due to the fast line speed and scarcity of fast memory in routers. In this work, we study the impact of allowing \emph{slack} in the window size on the asymptotic requirements of sliding window problems. That is, the algorithm can dynamically adjust the window size between $W$ and $W(1+τ)$ where $τ$ is a small positive parameter. We demonstrate this model's attractiveness by showing that it enables efficient algorithms to problems such as MAX and GENERAL-SUM that require $Ω(W)$ bits even for constant factor approximations in the exact sliding window model. Additionally, for problems that admit sub-linear approximation algorithms such as BASIC-SUMMING and COUNT-DISTINCT, the slack model enables a further asymptotic improvement.

cs.DS↗

arXiv2017

Pay for a Sliding Bloom Filter and Get Counting, Distinct Elements, and Entropy for Free

For many networking applications, recent data is more significant than older data, motivating the need for sliding window solutions. Various capabilities, such as DDoS detection and load balancing, require insights about multiple metrics including Bloom filters, per-flow counting, count distinct and entropy estimation. In this work, we present a unified construction that solves all the above problems in the sliding window model. Our single solution offers a better space to accuracy tradeoff than the state-of-the-art for each of these individual problems! We show this both analytically and by running multiple real Internet backbone and datacenter packet traces.

cs.DS↗

arXiv2017

Fast Flow Volume Estimation

The increasing popularity of jumbo frames means growing variance in the size of packets transmitted in modern networks. Consequently, network monitoring tools must maintain explicit traffic volume statistics rather than settle for packet counting as before. We present constant time algorithms for volume estimations in streams and sliding windows, which are faster than previous work. Our solutions are formally analyzed and are extensively evaluated over multiple real-world packet traces as well as synthetic ones. For streams, we demonstrate a run-time improvement of up to 2.4X compared to the state of the art. On sliding windows, we exhibit a memory reduction of over 100X on all traces and an asymptotic runtime improvement to a constant. Finally, we apply our approach to hierarchical heavy hitters and achieve an empirical 2.4-7X speedup.

cs.DS↗

arXiv2017

Constant Time Updates in Hierarchical Heavy Hitters

Monitoring tasks, such as anomaly and DDoS detection, require identifying frequent flow aggregates based on common IP prefixes. These are known as \emph{hierarchical heavy hitters} (HHH), where the hierarchy is determined based on the type of prefixes of interest in a given application. The per packet complexity of existing HHH algorithms is proportional to the size of the hierarchy, imposing significant overheads. In this paper, we propose a randomized constant time algorithm for HHH. We prove probabilistic precision bounds backed by an empirical evaluation. Using four real Internet packet traces, we demonstrate that our algorithm indeed obtains comparable accuracy and recall as previous works, while running up to 62 times faster. Finally, we extended Open vSwitch (OVS) with our algorithm and showed it is able to handle 13.8 million packets per second. In contrast, incorporating previous works in OVS only obtained 2.5 times lower throughput.

cs.DS↗

arXiv2017

Optimal Elephant Flow Detection

Monitoring the traffic volumes of elephant flows, including the total byte count per flow, is a fundamental capability for online network measurements. We present an asymptotically optimal algorithm for solving this problem in terms of both space and time complexity. This improves on previous approaches, which can only count the number of packets in constant time. We evaluate our work on real packet traces, demonstrating an up to X2.5 speedup compared to the best alternative.

cs.DS↗

arXiv2016

Randomized Admission Policy for Efficient Top-k and Frequency Estimation

Network management protocols often require timely and meaningful insight about per flow network traffic. This paper introduces Randomized Admission Policy (RAP) - a novel algorithm for the frequency and top-k estimation problems, which are fundamental in network monitoring. We demonstrate space reductions compared to the alternatives by a factor of up to 32 on real packet traces and up to 128 on heavy-tailed workloads. For top-k identification, RAP exhibits memory savings by a factor of between 4 and 64 depending on the skew of the workload. These empirical results are backed by formal analysis, indicating the asymptotic space improvement of our probabilistic admission approach. Additionally, we present d-Way RAP, a hardware friendly variant of RAP that empirically maintains its space and accuracy benefits.

cs.DS↗

arXiv2016

Hardening Cassandra Against Byzantine Failures

Cassandra is one of the most widely used distributed data stores these days. Cassandra supports flexible consistency guarantees over a wide-column data access model and provides almost linear scale-out performance. This enables application developers to tailor the performance and availability of Cassandra to their exact application's needs and required semantics. Yet, Cassandra is designed to withstand benign failures, and cannot cope with most forms of Byzantine attacks. In this work, we present an analysis of Cassandra's vulnerabilities and propose protocols for hardening Cassandra against Byzantine failures. We examine several alternative design choices and compare between them both qualitatively and empirically by using the Yahoo! Cloud Serving Benchmark (YCSB) performance benchmark. We include incremental performance analysis for our algorithmic and cryptographic adjustments, supporting our design choices.

cs.DC↗

arXiv2016

ICE Buckets: Improved Counter Estimation for Network Measurement

Measurement capabilities are essential for a variety of network applications, such as load balancing, routing, fairness and intrusion detection. These capabilities require large counter arrays in order to monitor the traffic of all network flows. While commodity SRAM memories are capable of operating at line speed, they are too small to accommodate large counter arrays. Previous works suggested estimators, which trade precision for reduced space. However, in order to accurately estimate the largest counter, these methods compromise the accuracy of the smaller counters. In this work, we present a closed form representation of the optimal estimation function. We then introduce Independent Counter Estimation Buckets (ICE-Buckets), a novel algorithm that improves estimation accuracy for all counters. This is achieved by separating the flows to buckets and configuring the optimal estimation function according to each bucket's counter scale. We prove a tighter upper bound on the relative error and demonstrate an accuracy improvement of up to 57 times on real Internet packet traces.

cs.NI↗

arXiv2016

COARA: Code Offloading on Android with AspectJ

Smartphones suffer from limited computational capabilities and battery life. A method to mitigate these problems is code offloading: executing application code on a remote server. We introduce COARA, a middleware platform for code offloading on Android that uses aspect-oriented programming (AOP) with AspectJ. AOP allows COARA to intercept code for offloading without a customized compiler or modification of the operating system. COARA requires minimal changes to application source code, and does not require the application developer to be aware of AOP. Since state transfer to the server is often a bottleneck that hinders performance, COARA uses AOP to intercept the transmission of large objects from the client and replaces them with object proxies. The server can begin execution of the offloaded application code, regardless of whether all required objects been transferred to the server. We run COARA with Android applications from the Google Play store on a Nexus 4 running unmodified Android 4.3 to prove that our platform improves performance and reduces energy consumption. Our approach yields speedups of 24x and 6x over WiFi and 3G respectively.

cs.DC↗

arXiv2016

Efficient Summing over Sliding Windows

This paper considers the problem of maintaining statistic aggregates over the last W elements of a data stream. First, the problem of counting the number of 1's in the last W bits of a binary stream is considered. A lower bound of Ω(1/ε + log W) memory bits for Wε-additive approximations is derived. This is followed by an algorithm whose memory consumption is O(1/ε + log W) bits, indicating that the algorithm is optimal and that the bound is tight. Next, the more general problem of maintaining a sum of the last W integers, each in the range of {0,1,...,R}, is addressed. The paper shows that approximating the sum within an additive error of RWε can also be done using Θ(1/ε + log W) bits for ε=Ω(1/W). For ε=o(1/W), we present a succinct algorithm which uses B(1 + o(1)) bits, where B=Θ(Wlog(1/Wε)) is the derived lower bound. We show that all lower bounds generalize to randomized algorithms as well. All algorithms process new elements and answer queries in O(1) worst-case time.

cs.DS↗