Ran Ben Basat

Ran Ben Basat

Auto-generated profile, based on publicly available information

Ran Ben Basat

University College London, London, United Kingdom

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
16 April 2020 in IEEE/ACM Transactions on Networking
Verified
Journal Article
Closed access
RBRan Ben Basat
Ran Ben Basat
XCXiaoqi Chen
Xiaoqi Chen
GEGil Einziger
Gil Einziger
OROri Rottenstreich
Ori Rottenstreich
Programmable network switches promise flexibility and high throughput, enabling applications such as load balancing and traffic engineering. Network measurement is a fundamental building block for such applications, including tasks such as the identification of heavy hitters (largest flows) or the detection of traffic changes. However, high-throughput packet processing architectures place certain limitations on the programming model, such as restricted branching, limited capability for memory access, and a limited number of processing stages. These limitations restrict the types of measurement algorithms that can run on programmable switches. In this paper, we focus on the Reconfigurable Match Tables (RMT) programmable high-throughput switch architecture, and carefully examine its constraints on designing measurement algorithms. We demonstrate our findings while solving the heavy hitter problem. We introduce PRECISION, an algorithm that uses Partial Recirculation to find top flows on a programmable switch. By recirculating a small fraction of packets, PRECISION simplifies the access to stateful memory to conform with RMT limitations and achieves higher accuracy than previous heavy hitter detection algorithms that avoid recirculation. We also evaluate each of the adaptations made by PRECISION and analyze its effect on the measurement accuracy. Finally, we suggest two algorithms for the hierarchical heavy hitters detection problem in which the goal is identifying the subnets that send excessive traffic and are potentially malicious. To the best of our knowledge, our work is the first to do so on RMT switches.

10 September 2023
Verified
Conference Paper
Open access
JLJonatan Langlet
Jonatan Langlet
RBRan Ben Basat
Ran Ben Basat
GOGabriele Oliaro
Gabriele Oliaro
MMMichael Mitzenmacher
Michael Mitzenmacher
MYMinlan Yu
Minlan Yu
GAGianni Antichi
Gianni Antichi
Fine-grained network telemetry is becoming a modern datacenter standard and is the basis of essential applications such as congestion control, load balancing, and advanced troubleshooting. As network size increases and telemetry gets more fine-grained, there is a tremendous growth in the amount of data needed to be reported from switches to collectors to enable network-wide view. As a consequence, it is progressively hard to scale data collection systems. We introduce Direct Telemetry Access (DTA), a solution optimized for aggregating and moving hundreds of millions of reports per second from switches into queryable data structures in collectors' memory. DTA is lightweight and it is able to greatly reduce overheads at collectors. DTA is built on top of RDMA, and we propose novel and expressive reporting primitives to allow easy integration with existing state-of-the-art telemetry mechanisms such as INT or Marple. We show that DTA significantly improves telemetry collection rates. For example, when used with INT, it can collect and aggregate over 400M reports per second with a single server, improving over the Atomic MultiLog by up to 16x.

26 May 2023 in Proceedings of the ACM on Management of Data
Verified
Journal Article
Open access
RSRana Shahout
Rana Shahout
RFRoy Friedman
Roy Friedman
RBRan Ben Basat
Ran Ben Basat
Stream monitoring is fundamental in many data stream applications, such as financial data trackers, security, anomaly detection, and load balancing. In that respect, quantiles are of particular interest, as they often capture the user's utility. For example, if a video connection has high tail (e.g., 99'th percentile) latency, the perceived quality will suffer, even if the average and median latencies are low. In this work, we consider the problem of approximating the per-item quantiles. Elements in our stream are (ID, value) tuples, and we wish to track the quantiles for each ID. Existing quantile sketches are designed for a plain number stream (i.e., containing just a value). While one could allocate a separate sketch instance for each ID, this may require an infeasible amount of memory. Instead, we consider tracking the quantiles for the heavy hitters (most frequent items), which are often considered particularly important, without knowing them beforehand. We first present a couple of simple and effective algorithms that serve as baselines, a sampling approach and a sketching approach. Then, we present SQUAD, an algorithm that combines sampling and sketching while improving the asymptotic space complexity. Intuitively, SQUAD uses a background sampling process to capture the behaviour of the quantiles of an item before it is allocated with a sketch, thereby allowing us to use fewer samples and sketches. The algorithms are rigorously analyzed, and we demonstrate SQUAD's superiority using extensive~simulations on real-world traces.

18 November 2024
Verified
Conference Paper
Open access
XCXiaoqi Chen
Xiaoqi Chen
SVShay Vargaftik
Shay Vargaftik
RBRan Ben Basat
Ran Ben Basat
Distributed training of ML models generates significant network traffic when exchanging gradients and is sensitive to packet drops and retransmission caused by congestion when other traffic is sharing the network. While training processes can tolerate gradient compression to reduce traffic volume, it does not eliminate queue buildup in shallow-buffer switches, causing high latency and extended training time. Our observation is that the emerging capability of handling temporal congestion through packet trimming provides the perfect opportunity to enable just-in-time gradient compression. Accordingly, we propose a novel approach where gradient packets are specifically constructed to be compressible via packet trimming, thus avoiding retransmission delays. Preliminary evaluations indicate that our trimmable encoding based on the Randomized Hadamard Transform has low computational overhead and that it achieves good training quality and shorter time to accuracy at high trim rates of up to 50%.

25 November 2024 in Proceedings of the ACM on Networking
Verified
Journal Article
Open access
WFWeiqi Feng
Weiqi Feng
JGJing Gao
Jing Gao
XCXiaoqi Chen
Xiaoqi Chen
GAGianni Antichi
Gianni Antichi
RBRan Ben Basat
Ran Ben Basat
MYMinlan Yu
Minlan Yu
Traffic monitoring in the dataplane is vital for reacting to network events such as microbursts, incast, and attacks. However, current solutions are constrained by the limited resources available on modern ASICs and don't provide the flexibility required to identify repeating patterns, such as applications whose flows communicate with a server at regular intervals. While such flexibility can be achieved using a co-processing CPU, it is generally too slow to provide insights quickly enough. In this paper, we show how an FPGA co-processor placed alongside the switching pipeline enables flexible traffic monitoring at data plane rates. While FPGAs have large memory and expressive processing, their throughput is significantly lower than switch ASICs. To bridge the throughput gap, we split query execution between the switch and FPGA and present methods that prevent processing all packets in FPGA. As a result, our system misses up to 5.0x fewer DDoS attack vectors than ACC-Turbo, the state-of-the-art on-switch solution, and up to 24% fewer microburst-contributing flows for the same precision rate.

18 November 2024
Verified
Conference Paper
Open access
WHWenchen Han
Wenchen Han
SVShay Vargaftik
Shay Vargaftik
MMMichael Mitzenmacher
Michael Mitzenmacher
BKBrad Karp
Brad Karp
RBRan Ben Basat
Ran Ben Basat
Gradient aggregation has long been identified as a major bottleneck in today's large-scale distributed machine learning training systems. One promising solution to mitigate such bottlenecks is gradient compression, directly reducing communicated gradient data volume. However, in practice, many gradient compression schemes do not achieve acceleration of the training process while also preserving accuracy. In this work, we identify common issues in previous gradient compression systems and evaluation methodologies. These include excessive computational overheads; incompatibility with all-reduce; and insufficient evaluation methods, such as not using an end-to-end metric or using a 32-bit baseline instead of the stronger 16-bit baseline. We revisit common compression approaches (sparsification, quantization, and low-rank decomposition) and demonstrate how considering the above issues can lead to minor but strategic design changes, resulting in notably better performance. Our goal is to raise awareness of the need for design and evaluation standards that naturally translate to the end-to-end utility of gradient compression.