We present a decade of evolution and production experience with Jupiter datacenter network fabrics. In this period Jupiter has delivered 5x higher speed and capacity, 30% reduction in capex, 41% reduction in power, incremental deployment and technology refresh all while serving live production traffic. A key enabler for these improvements is evolving Jupiter from a Clos to a direct-connect topology among the machine aggregation blocks. Critical architectural changes for this include: A datacenter interconnection layer employing Micro-Electro-Mechanical Systems (MEMS) based Optical Circuit Switches (OCSes) to enable dynamic topology reconfiguration, centralized Software-Defined Networking (SDN) control for traffic engineering, and automated network operations for incremental capacity delivery and topology engineering. We show that the combination of traffic and topology engineering on direct-connect fabrics achieves similar throughput as Clos fabrics for our production traffic patterns. We also optimize for path lengths: 60% of the traffic takes direct path from source to destination aggregation blocks, while the remaining transits one additional block, achieving an average block-level path length of 1.4 in our fleet today. OCS also achieves 3x faster fabric reconfiguration compared to pre-evolution Clos fabrics that used a patch panel based interconnect.

We present a decade of evolution and production experience with Jupiter datacenter network fabrics. In this period Jupiter has delivered 5x higher speed and capacity, 30% reduction in capex, 41% reduction in power, incremental deployment and technology refresh all while serving live production traffic. A key enabler for these improvements is evolving Jupiter from a Clos to a direct-connect topology among the machine aggregation blocks . Critical architectural changes for this include: A datacenter interconnection layer employing Micro-Electro-Mechanical Systems (MEMS) based Optical Circuit Switches (OCSes) to enable dynamic topology reconfiguration, centralized Software-Defined Networking (SDN) control for traffic engineering, and automated network operations for incremental capacity delivery and topology engineering. We show that the combination of traffic and topology engineering on direct-connect fabrics achieves similar throughput as Clos fabrics for our production traffic patterns. We also optimize for path lengths: 60% of the traffic takes direct path from source to destination aggregation blocks, while the remaining transits one additional block, achieving an average block-level path length of 1.4 in our fleet today. OCS also achieves 3x faster fabric reconfiguration compared to pre-evolution Clos fabrics that used a patch panel based interconnect.

With recent performance increase in the area of wireless mobile communications, mobile ad-hoc networks are playing a wide spread usage in the areas of military and other applications. But this mobile ad-hoc network does not have any centralized authorities like an access-point or a router as in case of wireless and wired networks to control and take care of routing. Thus routing has become a greater challenge to these types of networks. This paper proposes a new reputation based routing protocol based on DSR (Dynamic Source Routing) and through simulation results proves that the proposed method performs well compared to normal DSR.

ion (with a SQL-like language) is too high-level; how would languages like Erlang fit into this project? In fact, the authors really thought about using a functional language instead of a custom DSL. Someone asked whether they could still run into inconsistencies between layers. Tim replied that they don’t allow state to expire due to lack of use. They are adding the ability to have timeout events for removing rules. Aditya Akella (University of Wisconsin–Madison) asked about how the compiler works and whether it can handle all types of consistency semantics. Tim replied that Flowlog supports some things that OpenFlow does not. When their compiler detects forwarding rules, it produces netcore code and uses the netcore compiler. Andrew Ferguson (another author) pointed out that some parts of the policy might be directed toward a single switch, and that that is an abstraction introduced by netcore.

The potential advantages of optics at high link speeds have led to significant interest in deploying optical switching technology in data-center networks. Initial efforts have focused on hybrid approaches that rely on millisecond-scale circuit switching in the core of the network, while maintaining the flexibility of electrical packet switching at the edge. Recent demonstrations of microsecond-scale optical circuit switches motivate considering circuit switching for more dynamic traffic such as that generated from a top-of-rack (ToR) switch. Based on these technology trends, we propose a prototype hybrid ToR, called REACToR, which utilizes a combination of packet switching and circuit switching to appear to end-hosts as a packet-switched ToR. In this paper, we describe a prototype REACToR control plane which synchronizes end host transmissions with end-to-end circuit assignments. This control plane can react to rapid, bursty changes in the traffic from end hosts on a time scale of 100s of microseconds, several orders of magnitude faster than previous hybrid approaches. Using the experimental data from a system of eight end hosts, we calibrate a hybrid network simulator and use this simulator to predict the performance of larger-scale hybrid networks.

This paper proposes a combination of an Intrusion Detection System with a routing protocol to strengthen the defense of a Mobile Ad hoc Network. Our system is Socially Inspired, since we use the new paradigm of Reputation inherited from human behavior. The proposed IDS also has a unique characteristic of being Semi-distributed, since it neither distributes its Observation results globally nor keeps them entirely locally; however, managing to communicate this vital information without accretion of the network traffic. This innovative approach also avoids void assumptions and complex calculations for calculating and maintaining trust values used to estimate the reliability of other nodes observations. A robust Path Manager and Monitor system and Redemption and Fading concepts are other salient features of this design. The design has shown to outperform normal DSR in terms of Packet Delivery Ratio and Routing Overhead even when up to half of nodes in the network behave as malicious.

Solving “Big Data” problems requires bridging massive quantities of compute, memory, and storage, which requires a very high bandwidth network. Recently proposed direct connect networks like HyperX [1] and Flattened Butterfly [20] offer large capacity through paths of varying lengths between servers, and are highly cost effective for common data center workloads. However data center deployments are constrained to multi-rooted tree topologies like Fat-tree [2] and VL2 [16] due to shortest path routing and the limitations of commodity data center switch silicon. In this work we present Dahu1, simple enhancements to commodity Ethernet switches to support direct connect networks in data centers. Dahu avoids congestion hot-spots by dynamically spreading traffic uniformly across links, and forwarding traffic over non-minimal paths where possible. By performing load balancing primarily using local information, Dahu can act more quickly than centralized approaches, and responds to failure gracefully. Our evaluation shows that Dahu delivers up to 500% improvement in throughput over ECMP in large scale HyperX networks with over 130,000 servers, and up to 50% higher throughput in an 8,192 server Fat-tree network.

We are designing REACToR, a hybrid ToR that enables scalable, high-speed networks by pairing optical circuit switching with comparatively underprovisioned packet switching. The key insight driving our work is that by moving the vast majority-but not all-of the buffering out of the switch and into end hosts, more scalable, lower-cost interconnect fabrics can be supported.

While numerous studies have examined the macro-level behavior of traffic in data center networks---overall flow sizes, destination variability, and TCP burstiness---little information is available on the behavior of data center traffic at packet-level timescales. Whereas one might assume that flows from different applications fairly share available link bandwidth, and that packets within a single flow are uniformly paced, the reality is more complex. To meet increasingly high link rates of 10 Gbps and beyond, batching is typically introduced across the network stack---at the application, middleware, OS, transport, and NIC layers. This batching results in short-term packet bursts, which have implications for the design and performance requirements of packet processing devices along the path, including middleboxes, SDN-enabled switches, and virtual machine hypervisors. In this paper, we study the burst behavior of traffic emanating from a 10-Gbps end host across a variety of data center applications. We find that at 10--100 microsecond timescales, the traffic exhibits large bursts (i.e., 10s of packets in length). We further find that this level of burstiness is largely outside of application control, and independent of the high-level behavior of applications.

Engineering large-scale data center applications built from thousands of commodity nodes requires both an underlying network that supports a wide variety of traffic demands, and low latency at microsecond timescales. Many ideas for adding innovative functionality to networks, especially active queue management strategies, require either modifying packets or performing alternative queuing to packets in-flight on the data plane. However, configuring packet queuing, marking, and dropping is challenging, since buffering in commercial switches and routers is not programmable.

"Big Data" computing increasingly utilizes the MapReduce programming model for scalable processing of large data collections. Many MapReduce jobs are I/O-bound, and so minimizing the number of I/O operations is critical to improving their performance. In this work, we present Themis, a MapReduce implementation that reads and writes data records to disk exactly twice, which is the minimum amount possible for data sets that cannot fit in memory. In order to minimize I/O, Themis makes fundamentally different design decisions from previous MapReduce implementations. Themis performs a wide variety of MapReduce jobs -- including click log analysis, DNA read sequence alignment, and PageRank -- at nearly the speed of TritonSort's record-setting sort performance [29].

Engineering large-scale data center applications built from thousands of commodity nodes requires both an underlying network that supports a wide variety of traffic demands, and low latency at microsecond timescales. Many ideas for adding innovative functionality to networks, especially active queue management strategies, require either modifying packets or performing alternative queuing to packets in-flight on the data plane. However, configuring packet queuing, marking, and dropping is challenging, since buffering in commercial switches and routers is not programmable.In this work, we present NetBump, a platform for experimenting with, evaluating, and deploying a wide variety of active queue management strategies to network data planes with minimal intrusiveness and at low latency. NetBump leaves existing switches and endhosts unmodified by acting as a "bump on the wire,"examining, marking, and forwarding packets at line rate in tens ofmicroseconds to implement a variety of virtual active queuing disciplines and congestion control mechanisms. We describe the design of NetBump, and use it to implement several network functions and congestion control protocols including DCTCP and 802.1Qau quantized congestion notification.

Storage for cluster applications is typically provisioned based on rough, qualitative characterizations of applications. Moreover, configurations are often selected based on rules of thumb and are usually homogeneous across a deployment; to handle increased load, the application is simply scaled out across additional machines and storage of the same type. As deployments grow larger and storage options (e.g., disks, SSDs, DRAM) diversify, however, current practices are becoming increasingly inefficient in trading off cost versus performance. To enable more cost-effective deployment of cluster applications, we develop scc--a storage configuration compiler for cluster applications. scc automates cluster configuration decisions based on formal specifications of application behavior and hardware properties. We study a range of storage configurations and identify specifications that succinctly capture the trade-offs offered by different types of hardware, as well as the varying demands of application components. We apply scc to three representative applications and find that scc is expressive enough to meet application Service Level Agreements (SLAs) while delivering 2-4.5× savings in cost on average compared to simple scale-out options. scc's advantage stems mainly from its ability to configure heterogeneous--rather than conventional, homogeneous--cluster architectures to optimize cost.

In data center applications, predictability in service time and controlled latency, especially tail latency, are essential for building performant applications. This is especially true for applications or services built by accessing data across thousands of servers to generate a user response. Current practice has been to run such services at low utilization to rein in latency outliers, which decreases efficiency and limits the number of service invocations developers can issue while still meeting tight latency budgets. In this paper, we analyze three data center applications, Memcached, OpenFlow, and Web search, to measure the effect of 1) kernel socket handling, NIC interaction, and the network stack, 2) application locks contested in the kernel, and 3) application-layer queueing due to requests being stalled behind straggler threads on tail latency. We propose Chronos, a framework to deliver predictable, low latency in data center applications. Chronos uses a combination of existing and new techniques to achieve this end, for example by supporting Memcached at 200,000 requests per second per server at mean latency of 10 μs with a 99th percentile latency of only 30 μs, a factor of 20 lower than baseline Memcached.

This paper presents the design and implementation of an incrementally scalable architecture for middleboxes based on commodity servers and operating systems. xOMB, the eXtensible Open MiddleBox, employs general programmable network processing pipelines, with user-defined C++ modules responsible for parsing, transforming, and forwarding network flows. We implement three processing pipelines in xOMB, demonstrating good performance for load balancing, protocol acceleration, and application integration. In particular, our xOMB load balancing switch is able to match or outperform a commercial programmable switch and popular open-source reverse proxy while still providing a more flexible programming model.