Copyright notice for published papers: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author’s copyright.
Copyright notice for presentations and other work: All of my presentations, talks, and images on this page are available with Creative Commons Attribution ShareAlike 4.0 license.
Kumar, Dhruv; Wolfrath, Joel; Chandra, Abhishek; Sitaraman, Ramesh K.
In: In 5th ACM International Workshop on Edge Systems, Analytics and Networking (EdgeSys’22), 2022.
Large scale data analytics over geographically distributed data sources is challenging primarily due to the constrained and hetero- geneous resource availability such as the wide area network (WAN) bandwidth. In this work, we look at the problem of generating random samples over joins for geo-distributed data sources. Joins are one of the most fundamental yet expensive operations in data analytics. To reduce the cost of computing joins, existing techniques have looked at efficiently generating a random sample over the join result for centralized environments, where all the data is available in one location. These techniques fail to address the unique chal- lenges posed by geo-distributed environments. To address these challenges, we propose a sampling technique which aims to reduce the WAN traffic and latency, thereby reducing the overall latency for generating samples over joins for geo-distributed data sources. We implement our geo-distributed sampling technique on top of Apache Spark and compare it with existing state-of-the-art sam- pling techniques to identify scenarios where the proposed approach gives significant benefits. Based on this exploration, we provide a detailed outline of additional factors which should be consid- ered when designing a WAN-aware join sampling technique for geo-distributed environments.
Yang, Juncheng; Sabnis, Anirudh; Berger, Daniel S.; Rashmi, K. V.; Sitaraman, Ramesh K.
In: 19th USENIX Symposium on Networked Systems Design and Implementation (NSDI 22), pp. 1159–1177, USENIX Association, Renton, WA, 2022, ISBN: 978-1-939133-27-4.
Content Delivery Networks (CDNs) deliver much of the world’s web and video content to users from thousands of clusters deployed at the “edges” of the Internet. Maintain- ing consistent performance in this large distributed system is challenging. Through analysis of month-long logs from over 2000 clusters of a large CDN, we study the patterns of server unavailability. For a CDN with no redundancy, each server unavailability causes a sudden loss in performance as the objects previously cached on that server are not accessible, which leads to a miss ratio spike. The state-of-the-art miti- gation technique used by large CDNs is to replicate objects across multiple servers within a cluster. We find that although replication reduces miss ratio spikes, spikes remain a perfor- mance challenge. We present C2DN, the first CDN design that achieves a lower miss ratio, higher availability, higher resource efficiency, and close-to-perfect write load balancing. The core of our design is to introduce erasure coding into the CDN architecture and use the parity chunks to re-balance the write load across servers. We implement C2DN on top of open-source production software and demonstrate that com- pared to replication-based CDNs, C2DN obtains 11% lower byte miss ratio, eliminates unavailability-induced miss ratio spikes, and reduces write load imbalance by 99%.
Yang, Lin; Zeynali, Ali; Hajiesmaili, Mohammad H.; Sitaraman, Ramesh K.; Towsley, Don
In: Proceedings of the ACM on the Measurement Analysis of Computing Systems (POMACS), 2021.
Kumar, Dhruv; Ahmad, Sohaib; Chandra, Abhishek; Sitaraman, Ramesh K.
ACM/IEEE Symposium on Edge Computing (SEC '21), 2021.
Sabnis, Anirudh; Sitaraman, Ramesh
ACM Internet Measurement Conference (IMC), 2021.
Wang, Lingdong; Hajiesmaili, Mohammad; Sitaraman, Ramesh K
ACM International Conference on Multimedia (MM ’21), 2021.
Super-resolution (SR) is a well-studied technique for reconstructing high-resolu- tion (HR) images from low-resolution (LR) ones. SR holds great promise for video streaming since an LR video segment can be transmitted from the video server to the client that then reconstructs the HR version using SR, resulting in a significant reduction in network bandwidth. However, SR is seldom used in practice for real-time video streaming, because the computational overhead of frame reconstruction results in large latency and low frame rate.
To reduce the computational overhead and make SR practi- cal, we propose a deep-learning-based SR method called Foveated Cascaded Video Super Resolution (FOCAS). FOCAS relies on the fact that human eyes only have high acuity in a tiny central foveal region of the retina. FOCAS uses more neural network blocks in the foveal region to provide higher video quality, while using fewer blocks in the periphery as lower quality is sufficient. To optimize the computational resources and reduce reconstruction latency, FOCAS formulates and solves a convex optimization problem to decide the number of neural network blocks to use in each region of the frame. Using extensive experiments, we show that FOCAS reduces the latency by 50% − 70% while maintaining comparable visual quality as traditional (non-foveated) SR. Further, FOCAS provides a 12 − 16× reduction in the client-to-server network bandwidth in comparison with sending the full HR video segments.
Zhang, Xiao; Sen, Tanmoy; Zhang, Zheyuan; April, Tim; Chandrasekaran, Balakrishnan; Choffnes, David; Maggs, Bruce M.; Shen, Haiying; Sitaraman, Ramesh K.; Yang, Xiaowei
AnyOpt: Predicting and Optimizing IP Anycast Performance Inproceedings
In: Proceedings of the ACM SIGCOMM Conference, 2021.
Lee, Russell; Maghakian, Jessica; Hajiesmaili, Mohammad; Li, Jian; Sitaraman, Ramesh; Liu, Zhenhua
ACM International Conference on Future Energy Systems (ACM e-Energy), 2021.
Sabnis, Anirudh; Salem, Tareq Si; Neglia, Giovanni; Garetto, Michele; Leonardi, Emilio; Sitaraman, Ramesh K
GRADES: Gradient Descent for Similarity Caching Inproceedings
In: IEEE Conference on Computer Communications (INFOCOM), 2021.
Park, Jounsup; Wu, Mingyuan; Lee, Eric; Chen, Bo; Nahrstedt, Klara; Zink, Michael; Sitaraman, Ramesh
IEEE International Symposium on Multimedia (ISM), 2020.
Future view prediction for a 360-degree video streaming system is important to save the network bandwidth and improve the Quality of Experience (QoE). Historical view data of a single viewer and multiple viewers have been used for future view prediction. Video semantic information is also useful to predict the viewer’s future behavior. However, extracting video semantic information requires powerful computing hardware and large memory space to perform deep learning-based video analysis. It is not a desirable condition for most of client devices, such as small mobile devices or Head Mounted Display (HMD). Therefore, we develop an approach where video semantic analysis is executed on the media server, and the analysis results are shared with clients via the Semantic Flow Descriptor (SFD) and View-Object State Machine (VOSM). SFD and VOSM become new descriptive additions of the Media Presentation Description (MPD) and Spatial Relation Description (SRD) to support 360- degree video streaming via the DASH framework. Using the semantic-based approach, we design the Semantic-Aware View Prediction System (SEAWARE) to improve the overall view prediction performance. The evaluation results of 360-degree videos and real HMD view traces show that the SEAWARE system improves the view prediction performance and streams high-quality video with limited network bandwidth.
Liu, Yunzhuo; Jiang, Bo; Guo, Tian; Sitaraman, Ramesh; Towsley, Don; Wang, Xinbing
In: ACM Multimedia (MM) Conference, 2020.
Park, Jounsup; Shah, Yash; Rosenthal, Arielle; Wu, Mingyuan; Murray, John; Lee, Kuan-Ying; Spiteri, Kevin; Zink, Michael; Nahrstedt, Klara; Sitaraman, Ramesh
ACM Multimedia (MM) Conference, 2020.
Schomp, Kyle; Bhardwaj, Onkar; Kurdoglu, Eymen; Muhaimen, Mashooq; Sitaraman, Ramesh K.
In: Proceedings of the ACM SIGCOMM Conference, 2020.
We present Akamai DNS, one of the largest authoritative DNS infrastructures in the world, that supports the Akamai content de- livery network (CDN) as well as authoritative DNS hosting and DNS-based load balancing services for many enterprises. As the starting point for a significant fraction of the world’s Internet in- teractions, Akamai DNS serves millions of queries each second and must be resilient to avoid disrupting myriad online services, scalable to meet the ever increasing volume of DNS queries, per- formant to prevent user-perceivable performance degradation, and reconfigurable to react quickly to shifts in network conditions and attacks. We outline the design principles and architecture used to achieve Akamai DNS’s goals, relating the design choices to the system workload and quantifying the effectiveness of those designs. Further, we convey insights from operating the production system that are of value to the broader research community.
Sundarrajan, Aditya; Kasbekar, Mangesh; Sitaraman, Ramesh K; Shukla, Samta
Midgress-aware traffic provisioning for content delivery Inproceedings
In: USENIX Annual Technical Conference (USENIX ATC 20), pp. 543–557, USENIX Association, 2020, ISBN: 978-1-939133-14-4.
Content delivery networks (CDNs) cache and deliver hundreds of trillions of user requests each day from hundreds of thousands of servers around the world. The traffic served by CDNs can be partitioned into hundreds of traffic classes, each with different user access patterns, popularity distributions, object sizes, and performance requirements. Midgress is the cache miss traffic between the CDN's servers and the content provider origins. A major goal of a CDN is to minimize its midgress, since higher midgress translates to higher bandwidth costs and increased user-perceived latency. We propose algorithms that provision traffic classes to servers such that midgress is minimized. Using extensive traces from Akamai's CDN, we show that our midgress-aware traffic provisioning schemes can reduce midgress by nearly 20% in comparison with the midgress-unaware schemes currently in use. We also propose an efficient heuristic for traffic provisioning that achieves near-optimal midgress and is suitable for use in production settings. Further, we show how our algorithms can be extended to other settings that require minimum caching performance per traffic class and minimum content duplication for fault tolerance. Finally, our paper provides a strong case for implementing midgress-aware traffic provisioning in production CDNs.
Kirilin, Vadim; Sundarrajan, Aditya; Gorinsky, Sergey; Sitaraman, Ramesh K.
RL-Cache: Learning-Based Cache Admission for Content Delivery Journal Article
In: IEEE Journal on Selected Areas in Communications (JSAC), Special issue ssue on AI/ML for Networking and Communications, 2020.
Aditya Sundarrajan Vadim Kirilin, Sergey Gorinsky; Sitaraman, Ramesh K.
RL-Cache: Learning-Based Cache Admission for Content Delivery Journal Article
In: IEEE Journal on Selected Areas of Communications (JSAC), 2020.
Content delivery networks (CDNs) distribute much of the Internet content by caching and serving the objects requested by users. A major goal of a CDN is to maximize the hit rates of its caches, thereby enabling faster content downloads to the users. Content caching involves two components: an admission algorithm to decide whether to cache an object and an eviction algorithm to determine which object to evict from the cache when it is full. In this paper, we focus on cache admission and propose a novel algorithm called RL-Cache that uses model-free reinforcement learning (RL) to decide whether or not to admit a requested object into the CDN’s cache. Unlike prior approaches that use a small set of criteria for decision making, RL-Cache weights a large set of features that include the object size, recency, and frequency of access. We develop a publicly available implementation of RL-Cache and perform an evaluation using production traces for the image, video, and web traffic classes from Akamai’s CDN. The evaluation shows that RL-Cache improves the hit rate in comparison with the state of the art and imposes only a modest resource overhead on the CDN servers. Further, RL-Cache is robust enough that it can be trained in one location and executed on request traces of the same or different traffic classes in other locations of the same geographic region. The paper also reports extensive analyses of the RL-Cache sensitivity to its features and hyperparameter values. The analyses validate the made design choices and reveal interesting insights into the RL-Cache behavior.
Rahul Urgaonkar Kevin Spiteri, Ramesh Sitaraman
BOLA: Near-Optimal Bitrate Adaptation for Online Videos Journal Article
In: IEEE/ACM Transactions on Networking, 2020.
Mohammad H. Hajiesmaili Lin Yang, Ramesh Sitaraman; Wong, Wing S.
Online Linear Optimization with Inventory Management Constraints Journal Article
In: Proceedings of the ACM on Measurement and Analysis of Computing Systems, vol. 4, no. 1, pp. 29, 2020.
This paper considers the problem of online linear optimization with inventory management constraints. Specifically, we consider an online scenario where a decision maker needs to satisfy her time-varying demand for some units of an asset, either from a market with a time-varying price or from her own inventory. In each time slot, the decision maker is presented a (linear) price and must immediately decide the amount to purchase for covering the demand and/or for storing in the inventory for future use. The inventory has a limited capacity and can be used to buy and store assets at low price and cover the demand when the price is high. The ultimate goal of the decision maker is to cover the demand at each time slot while minimizing the cost of buying assets from the market. We propose ARP, an online algorithm for linear programming with inventory constraints, and ARPRate, an extended version that handles rate constraints to/from the inventory. Both ARP and ARPRate achieve optimal competitive ratios, meaning that no other online algorithm can achieve a better theoretical guarantee. To illustrate the results, we use the proposed algorithms in a case study focused on energy procurement and storage management strategies for data centers.
Sabnis, Anirudh Anirudh; Sitaraman, Ramesh K; Towsley, Donald
OCCAM: An optimization based approach to network inference Journal Article
In: ACM SIGMETRICS Performance Evaluation Review, vol. 46, no. 2, pp. 36–38, 2019.
Spiteri, Kevin; Sitaraman, Ramesh; Sparacio, Daniel
In: ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM), vol. 15, no. 2s, pp. 67, 2019.