Cauligi S. Raghavendra

Mohammad Noormohammadpour, Cauligi S. Raghavendra

Datacenters provide cost-effective and flexible access to scalable compute and storage resources necessary for today's cloud computing needs. A typical datacenter is made up of thousands of servers connected with a large network and usually managed by one operator. To provide quality access to the variety of applications and services hosted on datacenters and maximize performance, it deems necessary to use datacenter networks effectively and efficiently. Datacenter traffic is often a mix of several classes with different priorities and requirements. This includes user-generated interactive traffic, traffic with deadlines, and long-running traffic. To this end, custom transport protocols and traffic management techniques have been developed to improve datacenter network performance. In this tutorial paper, we review the general architecture of datacenter networks, various topologies proposed for them, their traffic properties, general traffic control challenges in datacenters and general traffic control objectives. The purpose of this paper is to bring out the important characteristics of traffic control in datacenters and not to survey all existing solutions (as it is virtually impossible due to massive body of existing research). We hope to provide readers with a wide range of options and factors while considering a variety of traffic control mechanisms. We discuss various characteristics of datacenter traffic control including management schemes, transmission control, traffic shaping, prioritization, load balancing, multipathing, and traffic scheduling. Next, we point to several open challenges as well as new and interesting networking paradigms. At the end of this paper, we briefly review inter-datacenter networks that connect geographically dispersed datacenters which have been receiving increasing attention recently and pose interesting and novel research problems.

Mohammad Noormohammadpour, Cauligi S. Raghavendra, Srikanth Kandula et al.

Large inter-datacenter transfers are crucial for cloud service efficiency and are increasingly used by organizations that have dedicated wide area networks between datacenters. A recent work uses multicast forwarding trees to reduce the bandwidth needs and improve completion times of point-to-multipoint transfers. Using a single forwarding tree per transfer, however, leads to poor performance because the slowest receiver dictates the completion time for all receivers. Using multiple forwarding trees per transfer alleviates this concern--the average receiver could finish early; however, if done naively, bandwidth usage would also increase and it is apriori unclear how best to partition receivers, how to construct the multiple trees and how to determine the rate and schedule of flows on these trees. This paper presents QuickCast, a first solution to these problems. Using simulations on real-world network topologies, we see that QuickCast can speed up the average receiver's completion time by as much as $10\times$ while only using $1.04\times$ more bandwidth; further, the completion time for all receivers also improves by as much as $1.6\times$ faster at high loads.

Mohammad Noormohammadpour, Srikanth Kandula, Cauligi S. Raghavendra et al.

Bulk transfers from one to multiple datacenters can have many different completion time objectives ranging from quickly replicating some $k$ copies to minimizing the time by which the last destination receives a full replica. We design an SDN-style wide-area traffic scheduler that optimizes different completion time objectives for various requests. The scheduler builds, for each bulk transfer, one or more multicast forwarding trees which preferentially use lightly loaded network links. Multiple multicast trees are used per bulk transfer to insulate destinations that have higher available bandwidth and can hence finish quickly from congested destinations. These decisions--how many trees to construct and which receivers to serve using a given tree--result from an optimization problem that minimizes a weighted sum of transfers' completion time objectives and their bandwidth consumption. Results from simulations and emulations on Mininet show that our scheduler, Iris, can improve different completion time objectives by about $2.5\times$.

Mohammad Noormohammadpour, Cauligi S. Raghavendra

Inter-datacenter networks connect dozens of geographically dispersed datacenters and carry traffic flows with highly variable sizes and different classes. Adaptive flow routing can improve efficiency and performance by assigning paths to new flows according to network status and flow properties. A popular approach widely used for traffic engineering is based on current bandwidth utilization of links. We propose an alternative that reduces bandwidth usage by up to at least 50% and flow completion times by up to at least 40% across various scheduling policies and flow size distributions.

Mohammad Noormohammadpour, Ajitesh Srivastava, Cauligi S. Raghavendra

Long flows contribute huge volumes of traffic over inter-datacenter WAN. The Flow Completion Time (FCT) is a vital network performance metric that affects the running time of distributed applications and the users' quality of experience. Flow routing techniques based on propagation or queuing latency or instantaneous link utilization are insufficient for minimization of the long flows' FCT. We propose a routing approach that uses the remaining sizes and paths of all ongoing flows to minimize the worst-case completion time of incoming flows assuming no knowledge of future flow arrivals. Our approach can be formulated as an NP-Hard graph optimization problem. We propose BWRH, a heuristic to quickly generate an approximate solution. We evaluate BWRH against several real WAN topologies and two different traffic patterns. We see that BWRH provides solutions with an average optimality gap of less than $0.25\%$. Furthermore, we show that compared to other popular routing heuristics, BWRH reduces the mean and tail FCT by up to $1.46\times$ and $1.53\times$, respectively.

Ayush Jaiswal, Dong Guo, Cauligi S. Raghavendra et al.

Machine Learning (ML) is increasingly being used for computer aided diagnosis of brain related disorders based on structural magnetic resonance imaging (MRI) data. Most of such work employs biologically and medically meaningful hand-crafted features calculated from different regions of the brain. The construction of such highly specialized features requires a considerable amount of time, manual oversight and careful quality control to ensure the absence of errors in the computational process. Recent advances in Deep Representation Learning have shown great promise in extracting highly non-linear and information-rich features from data. In this paper, we present a novel large-scale deep unsupervised approach to learn generic feature representations of structural brain MRI scans, which requires no specialized domain knowledge or manual intervention. Our method produces low-dimensional representations of brain structure, which can be used to reconstruct brain images with very low error and exhibit performance comparable to FreeSurfer features on various classification tasks.

Mohammad Noormohammadpour, Cauligi S. Raghavendra, Sriram Rao

Datacenters provide the infrastructure for cloud computing services used by millions of users everyday. Many such services are distributed over multiple datacenters at geographically distant locations possibly in different continents. These datacenters are then connected through high speed WAN links over private or public networks. To perform data backups or data synchronization operations, many transfers take place over these networks that have to be completed before a deadline in order to provide necessary service guarantees to end users. Upon arrival of a transfer request, we would like the system to be able to decide whether such a request can be guaranteed successful delivery. If yes, it should provide us with transmission schedule in the shortest time possible. In addition, we would like to avoid packet reordering at the destination as it affects TCP performance. Previous work in this area either cannot guarantee that admitted transfers actually finish before the specified deadlines or use techniques that can result in packet reordering. In this paper, we propose DCRoute, a fast and efficient routing and traffic allocation technique that guarantees transfer completion before deadlines for admitted requests. It assigns each transfer a single path to avoid packet reordering. Through simulations, we show that DCRoute is at least 200 times faster than other traffic allocation techniques based on linear programming (LP) while admitting almost the same amount of traffic to the system.

Mohammad Noormohammadpour, Cauligi S. Raghavendra, Sriram Rao et al.

Using multiple datacenters allows for higher availability, load balancing and reduced latency to customers of cloud services. To distribute multiple copies of data, cloud providers depend on inter-datacenter WANs that ought to be used efficiently considering their limited capacity and the ever-increasing data demands. In this paper, we focus on applications that transfer objects from one datacenter to several datacenters over dedicated inter-datacenter networks. We present DCCast, a centralized Point to Multi-Point (P2MP) algorithm that uses forwarding trees to efficiently deliver an object from a source datacenter to required destination datacenters. With low computational overhead, DCCast selects forwarding trees that minimize bandwidth usage and balance load across all links. With simulation experiments on Google's GScale network, we show that DCCast can reduce total bandwidth usage and tail Transfer Completion Times (TCT) by up to $50\%$ compared to delivering the same objects via independent point-to-point (P2P) transfers.