Fast VMM-based overlay networking for bridging the cloud and high performance computing

A collection of virtual machines (VMs) interconnected with an overlay network with a layer 2 abstraction has proven to be a powerful, unifying abstraction for adaptive distributed and parallel computing on loosely-coupled environments. It is now feasible to allow VMs hosting high performance computing (HPC) applications to seamlessly bridge distributed cloud resources and tightly-coupled supercomputing and cluster resources. However, to achieve the application performance that the tightly-coupled resources are capable of, it is important that the overlay network not introduce significant overhead relative to the native hardware, which is not the case for current user-level tools, including our own existing VNET/U system. In response, we describe the design, implementation, and evaluation of a virtual networking system that has negligible latency and bandwidth overheads in 1–10 Gbps networks. Our system, VNET/P, is directly embedded into our publicly available Palacios virtual machine monitor (VMM). VNET/P achieves native performance on 1 Gbps Ethernet networks and very high performance on 10 Gbps Ethernet networks. The NAS benchmarks generally achieve over 95 % of their native performance on both 1 and 10 Gbps. We have further demonstrated that VNET/P can operate successfully over more specialized tightly-coupled networks, such as Infiniband and Cray Gemini. Our results suggest it is feasible to extend a software-based overlay network designed for computing at wide-area scales into tightly-coupled environments.     

Network slicing to improve multicasting in HPC clusters

In high performance computing (HPC) resources’ extensive experiments are frequently executed. HPC resources (e.g. computing machines and switches) should be able to handle running several experiments in parallel. Typically HPC utilizes parallelization in programs, processing and data. The underlying network is seen as the only non-parallelized HPC component (i.e. no dynamic virtual slicing based on HPC jobs). In this scope we present an approach in this paper to utilize software defined networking (SDN) to parallelize HPC clusters among the different running experiments. We propose to accomplish this through two major components: A passive module (network mapper/remapper) to select for each experiment as soon as it starts the least busy resources in the network, and an SDN-HPC active load balancer to perform more complex and intelligent operations. Active load balancer can logically divide the network based on experiments’ host files. The goal is to reduce traffic to unnecessary hosts or ports. An HPC experiment should multicast, rather than broadcast to only cluster nodes that are used by the experiment. We use virtual tenant network modules in Opendaylight controller to create VLANs based on HPC experiments. In each HPC host, virtual interfaces are created to isolate traffic from the different experiments. The traffic between the different physical hosts that belong to the same experiment can be distinguished based on the VLAN ID assigned to each experiment. We evaluate the new approach using several HPC public benchmarks. Results show a significant enhancement in experiments’ performance especially when HPC cluster experiences running several heavy load experiments simultaneously. Results show also that this multi-casting approach can significantly reduce casting overhead that is caused by using a single cast for all resources in the HPC cluster. In comparison with InfiniBand networks that offer interconnect services with low latency and high bandwidth, HPC services based on SDN can provide two distinguished objectives that may not be possible with InfiniBand: The first objective is the integration of HPC with Ethernet enterprise networks and hence expanding HPC usage to much wider domains. The second objective is the ability to enable users and their applications to customize HPC services with different QoS requirements that fit the different needs of those applications and optimize the usage of HPC clusters.     

FastMPJ: a scalable and efficient Java message-passing library

The performance and scalability of communications are key for high performance computing (HPC) applications in the current multi-core era. Despite the significant benefits (e.g., productivity, portability, multithreading) of Java for parallel programming, its poor communications support has hindered its adoption in the HPC community. This paper presents FastMPJ, an efficient message-passing in Java (MPJ) library, boosting Java for HPC by: (1) providing high-performance shared memory communications using Java threads; (2) taking full advantage of high-speed cluster networks (e.g., InfiniBand) to provide low-latency and high bandwidth communications; (3) including a scalable collective library with topology aware primitives, automatically selected at runtime; (4) avoiding Java data buffering overheads through zero-copy protocols; and (5) implementing the most widely extended MPI-like Java bindings for a highly productive development. The comprehensive performance evaluation on representative testbeds (InfiniBand, 10 Gigabit Ethernet, Myrinet, and shared memory systems) has shown that FastMPJ communication primitives rival native MPI implementations, significantly improving the efficiency and scalability of Java HPC parallel applications.     

Parallel application-level behavioral attributes for performance and energy management of high-performance computing systems

Run time variability of parallel applications continues to present significant challenges to their performance and energy efficiency in high-performance computing (HPC) systems. When run times are extended and unpredictable, application developers perceive this as a degradation of system (or subsystem) performance. Extended run times directly contribute to proportionally higher energy consumption, potentially negating efforts by applications, or the HPC system, to optimize energy consumption using low-level control techniques, such as dynamic voltage and frequency scaling (DVFS). Therefore, successful systemic management of application run time performance can result in less wasted energy, or even energy savings. We have been studying run time variability in terms of communication time, from the perspective of the application, focusing on the interconnection network. More recently, our focus has shifted to developing a more complete understanding of the effects of HPC subsystem interactions on parallel applications. In this context, the set of executing applications on the HPC system is treated as a subsystem, along with more traditional subsystems like the communication subsystem, storage subsystem, etc. To gain insight into the run time variability problem, our earlier work developed a framework to emulate parallel applications (PACE) that stresses the communication subsystem. Evaluation of run time sensitivity to network performance of real applications is performed with a tool called PARSE, which uses PACE. In this paper, we propose a model defining application-level behavioral attributes, that collectively describes how applications behave in terms of their run time performance, as functions of their process distribution on the system (spacial locality), and subsystem interactions (communication subsystem degradation). These subsystem interactions are produced when multiple applications execute concurrently on the same HPC system. We also revisit our evaluation framework and tools to demonstrate the flexibility of our application characterization techniques, and the ease with which attributes can be quantified. The validity of the model is demonstrated using our tools with several parallel benchmarks and application fragments. Results suggest that it is possible to articulate application-level behavioral attributes as a tuple of numeric values that describe course-grained performance behavior.     

记忆水平依赖的海马-前额叶神经节律交互

海马(HPC)和前额叶皮层(PFC)的协同作用是记忆加工过程的关键,其相互作用对学习和记忆功能至关重要.大量证据表明,情景记忆的形成、巩固与检索依赖于特征神经节律在PFC和HPC脑区间的同步作用,这些节律包括theta节律、gamma节律和sharp wave ripples (SWRs)节律等.在精神类疾病中患者往往伴随出现学习记忆功能障碍,基于人类和动物的脑电研究均发现以上3种神经节律在HPC和PFC之间的同步性下降,可能作为反映精神病理下认知功能障碍的重要指标.本文从HPC-PFC网络中的神经节律研究出发,总结了theta节律、gamma节律和SWRs节律在两脑区间的协调交互模式在情景记忆中的作用,以及精神分裂症和抑郁症状态下HPC-PFC通路上神经节律的异常表现及其潜在损伤机制,为今后精神疾病的快速诊断提供客观依据.     

Multi-Parametric Clustering for Sensor Node Coordination in Cognitive Wireless Sensor Networks

The deployment of wireless sensor networks for healthcare applications have been motivated and driven by the increasing demand for real-time monitoring of patients in hospital and large disaster response environments. A major challenge in developing such sensor networks is the need for coordinating a large number of randomly deployed sensor nodes. In this study, we propose a multi-parametric clustering scheme designed to aid in the coordination of sensor nodes within cognitive wireless sensor networks. In the proposed scheme, sensor nodes are clustered together based on similar network behaviour across multiple network parameters, such as channel availability, interference characteristics, and topological characteristics, followed by mechanisms for forming, joining and switching clusters. Extensive performance evaluation is conducted to study the impact on important factors such as clustering overhead, cluster joining estimation error, interference probability, as well as probability of reclustering. Results show that the proposed clustering scheme can be an excellent candidate for use in large scale cognitive wireless sensor network deployments with high dynamics.     

Adaptive resource utilization and remote access capabilities in high‐performance distributed systems: The Open HPC++ approach

Development of high-performance distributed applications, called metaapplications, is extremely challenging because of their complex runtime environment coupled with their requirements of high-performance and Quality of Service (QoS). Such applications typically run on a set of heterogeneous machines with dynamically varying loads, connected by heterogeneous networks possibly supporting a wide variety of communication protocols. In spite of the size and complexity of such applications, they must provide the high-performance and QoS mandated by their users. In order to achieve the goal of high-performance, they need to adaptively utilize their computational and communication resources. Apart from the requirements of adaptive resource utilization, such applications have a third kind of requirement related to remote access QoS. Different clients, although accessing a single server resource, may have differing QoS requirements from their remote connections. A single server resource may also need to provide different QoS for different clients, depending on various issues such as the amount of trust between the server and a given client. These QoS requirements can be encapsulated under the abstraction of remote access capabilities. Metaapplications need to address all the above three requirements in order to achieve the goal of high-performance and satisfy user expectations of QoS. This paper presents Open HPC++, a programming environment for high-performance applications running in a complex and heterogeneous run-time environment. Open HPC++ provides application level tools and mechanisms to satisfy application requirements of adaptive resource utilization and remote access capabilities. Open HPC++ is designed on the lines of CORBA and uses an Object Request Broker (ORB) to support seamless communication between distributed application components. In order to provide adaptive utilization of communication resources, it uses the principle of open implementation to open up the communication mechanisms of its ORB. By virtue of its open architecture, the ORB supports multiple, possibly custom, communication protocols, along with automatic and user controlled protocol selection at run-time. An extension of the same mechanism is used to support the concept of remote access capabilities. In order to support adaptive utilization of computational resources, Open HPC++ also provides a flexible yet powerful set of load-balancing mechanisms that can be used to implement custom load-balancing strategies. The paper also presents performance evaluations of Open HPC++ adaptivity and load-balancing mechanisms. This revised version was published online in July 2006 with corrections to the Cover Date.     

A Service Approach for Architecting Application Independent Wireless Sensor Networks

The current sensor networks are assumed to be designed for specific applications, having data communication protocols strongly coupled to applications. The future sensor networks are envisioned as comprising heterogeneous devices assisting to a large range of applications. To achieve this goal, a new architecture approach is needed, having application specific features separated from the data communication protocol, while influencing its behavior. We propose a Web Services approach for the design of sensor network, in which sensor nodes are service providers and applications are clients of such services. Our main goal is to enable a flexible architecture in which sensor networks data can be accessed by users spread all over the world.     

Unilateral Lateral Entorhinal Inactivation Impairs Memory Expression in Trace Eyeblink Conditioning

Memory in trace eyeblink conditioning is mediated by an inter-connected network that involves the hippocampus (HPC), several neocortical regions, and the cerebellum. This network reorganizes after learning as the center of the network shifts from the HPC to the medial prefrontal cortex (mPFC). Despite the network reorganization, the lateral entorhinal cortex (LEC) plays a stable role in expressing recently acquired HPC-dependent memory as well as remotely acquired mPFC-dependent memory. Entorhinal involvement in recent memory expression may be attributed to its previously proposed interactions with the HPC. In contrast, it remains unknown how the LEC participates in memory expression after the network disengages from the HPC. The present study tested the possibility that the LEC and mPFC functionally interact during remote memory expression by examining the impact of pharmacological inactivation of the LEC in one hemisphere and the mPFC in the contralateral hemisphere on memory expression in rats. Memory expression one day and one month after learning was significantly impaired after LEC-mPFC inactivation; however, the degree of impairment was comparable to that after unilateral LEC inactivation. Unilateral mPFC inactivation had no effect on recent or remote memory expression. These results suggest that the integrity of the LEC in both hemispheres is necessary for memory expression. Functional interactions between the LEC and mPFC should therefore be tested with an alternative design.     

Janus: a framework to boost HPC applications in the cloud based on SDN path provisioning

Data centers, clusters, and grids have historically supported High-Performance Computing (HPC) applications. Due to the high capital and operational expenditures associated with such infrastructures, we have witnessed consistent efforts to run HPC applications in the cloud in the recent past. The potential advantages of this shift include higher scalability and lower costs. If, on the one hand, app instantiation—through customized Virtual Machines (VMs)—is a well-solved issue, on the other, the network still represents a significant bottleneck. When switching HPC applications to be executed on the cloud, we lose control of where VMs will be positioned and of the paths that will be traversed for processes to communicate with one another. To bridge this gap, we present Janus, a framework for dynamic, just-in-time path provisioning in cloud infrastructures. By leveraging emerging software-defined networking principles, the framework allows for an HPC application, once deployed, to have interprocess communication paths configured upon usage based on least-used network links (instead of resorting to shortest, pre-computed paths). Janus is fully configurable to cope with different operating parameters and communication strategies, providing a rich ecosystem for application execution speed up. Through an extensive experimental evaluation, we provide evidence that the proposed framework can lead to significant gains regarding runtime. Moreover, we show what one can expect in terms of system overheads, providing essential insights on how better benefiting from Janus.

     

Performance evaluation of a remote memory system with commodity hardware for large-memory data processing

The explosion of data and transactions demands a creative approach for data processing in a variety of applications. Research on remote memory systems (RMSs), so as to exploit the superior characteristics of dynamic random access memory (DRAM), has been performed for many decades, and today’s information explosion galvanizes researchers into shedding new light on the technology. Prior studies have mainly focused on architectural suggestions for such systems, highlighting different design rationale. These studies have shown that choosing the appropriate applications to run on an RMS is important in fully utilizing the advantages of remote memory. This article provides an extensive performance evaluation for various types of data processing applications so as to address the efficacy of an RMS by means of a prototype RMS with reliability functionality. The prototype RMS used is a practical kernel-level RMS that renders large memory data processing feasible. The abstract concept of remote memory was materialized by borrowing unused local memory in commodity PCs via a high speed network capable of Remote Direct Memory Access (RDMA) operations. The prototype RMS uses remote memory without any part of its computation power coming from remote computers. Our experimental results suggest that an RMS can be practical in supporting the rigorous demands of commercial in memory database systems that have high data access locality. Our evaluation also convinces us of the possibility that a reliable RMS can satisfy both the high degree of reliability and efficiency for large memory data processing applications whose data access pattern has high locality.     

Exploiting temporal network structures of human interaction to effectively immunize populations

Decreasing the number of people who must be vaccinated to immunize a community against an infectious disease could both save resources and decrease outbreak sizes. A key to reaching such a lower threshold of immunization is to find and vaccinate people who, through their behavior, are more likely than average to become infected and to spread the disease further. Fortunately, the very behavior that makes these people important to vaccinate can help us to localize them. Earlier studies have shown that one can use previous contacts to find people that are central in static contact networks. However, real contact patterns are not static. In this paper, we investigate if there is additional information in the temporal contact structure for vaccination protocols to exploit. We answer this affirmative by proposing two immunization methods that exploit temporal correlations and showing that these methods outperform a benchmark static-network protocol in four empirical contact datasets under various epidemic scenarios. Both methods rely only on obtainable, local information, and can be implemented in practice. For the datasets directly related to contact patterns of potential disease spreading (of sexually-transmitted and nosocomial infections respectively), the most efficient protocol is to sample people at random and vaccinate their latest contacts. The network datasets are temporal, which enables us to make more realistic evaluations than earlier studies--we use only information about the past for the purpose of vaccination, and about the future to simulate disease outbreaks. Using analytically tractable models, we identify two temporal structures that explain how the protocols earn their efficiency in the empirical data. This paper is a first step towards real vaccination protocols that exploit temporal-network structure--future work is needed both to characterize the structure of real contact sequences and to devise immunization methods that exploit these.     

Suppression of neurotoxic lesion-induced seizure activity: evidence for a permanent role for the hippocampus in contextual memory

Damage to the hippocampus (HPC) using the excitotoxin N-methyl-D-aspartate (NMDA) can cause retrograde amnesia for contextual fear memory. This amnesia is typically attributed to loss of cells in the HPC. However, NMDA is also known to cause intense neuronal discharge (seizure activity) during the hours that follow its injection. These seizures may have detrimental effects on retrieval of memories. Here we evaluate the possibility that retrograde amnesia is due to NMDA-induced seizure activity or cell damage per se. To assess the effects of NMDA induced activity on contextual memory, we developed a lesion technique that utilizes the neurotoxic effects of NMDA while at the same time suppressing possible associated seizure activity. NMDA and tetrodotoxin (TTX), a sodium channel blocker, are simultaneously infused into the rat HPC, resulting in extensive bilateral damage to the HPC. TTX, co-infused with NMDA, suppresses propagation of seizure activity. Rats received pairings of a novel context with foot shock, after which they received NMDA-induced, TTX+NMDA-induced, or no damage to the HPC at a recent (24 hours) or remote (5 weeks) time point. After recovery, the rats were placed into the shock context and freezing was scored as an index of fear memory. Rats with an intact HPC exhibited robust memory for the aversive context at both time points, whereas rats that received NMDA or NMDA+TTX lesions showed a significant reduction in learned fear of equal magnitude at both the recent and remote time points. Therefore, it is unlikely that observed retrograde amnesia in contextual fear conditioning are due to disruption of non-HPC networks by propagated seizure activity. Moreover, the memory deficit observed at both time points offers additional evidence supporting the proposition that the HPC has a continuing role in maintaining contextual memories.     

Modeling,analysis and simulation of ant-based network routing protocols

Using the metaphor of swarm intelligence, ant-based routing protocols deploy control packets that behave like ants to discover and optimize routes between pairs of nodes. These ant-based routing protocols provide an elegant, scalable solution to the routing problem for both wired and mobile ad hoc networks. The routing problem is highly nonlinear because the control packets alter the local routing tables as they are routed through the network. We mathematically map the local rules by which the routing tables are altered to the dynamics of the entire networks. Using dynamical systems theory, we map local protocol rules to full network performance, which helps us understand the impact of protocol parameters on network performance. In this paper, we systematically derive and analyze global models for simple ant-based routing protocols using both pheromone deposition and evaporation. In particular, we develop a stochastic model by modeling the probability density of ants over the network. The model is validated by comparing equilibrium pheromone levels produced by the global analysis to results obtained from simulation studies. We use both a Matlab simulation with ideal communications and a QualNet simulation with realistic communication models. Using these analytic and computational methods, we map out a complete phase diagram of network behavior over a small multipath network. We show the existence of both stable and unstable (inaccessible) routing solutions having varying properties of efficiency and redundancy depending upon the routing parameters. Finally, we apply these techniques to a larger 50-node network and show that the design principles acquired from studying the small model network extend to larger networks.     

Learning, memory, and the role of neural network architecture

The performance of information processing systems, from artificial neural networks to natural neuronal ensembles, depends heavily on the underlying system architecture. In this study, we compare the performance of parallel and layered network architectures during sequential tasks that require both acquisition and retention of information, thereby identifying tradeoffs between learning and memory processes. During the task of supervised, sequential function approximation, networks produce and adapt representations of external information. Performance is evaluated by statistically analyzing the error in these representations while varying the initial network state, the structure of the external information, and the time given to learn the information. We link performance to complexity in network architecture by characterizing local error landscape curvature. We find that variations in error landscape structure give rise to tradeoffs in performance; these include the ability of the network to maximize accuracy versus minimize inaccuracy and produce specific versus generalizable representations of information. Parallel networks generate smooth error landscapes with deep, narrow minima, enabling them to find highly specific representations given sufficient time. While accurate, however, these representations are difficult to generalize. In contrast, layered networks generate rough error landscapes with a variety of local minima, allowing them to quickly find coarse representations. Although less accurate, these representations are easily adaptable. The presence of measurable performance tradeoffs in both layered and parallel networks has implications for understanding the behavior of a wide variety of natural and artificial learning systems.     

Short-term memory of motor network performance via activity-dependent potentiation of Na+/K+ pump function

Brain networks memorize previous performance to adjust their output in light of past experience. These activity-dependent modifications generally result from changes in synaptic strengths or ionic conductances, and ion pumps have only rarely been demonstrated to play a dynamic role. Locomotor behavior is produced by central pattern generator (CPG) networks and modified by sensory and descending signals to allow for changes in movement frequency, intensity, and duration, but whether or how the CPG networks recall recent activity is largely unknown. In Xenopus frog tadpoles, swim bout duration correlates linearly with interswim interval, suggesting that the locomotor network retains a short-term memory of previous output. We discovered an ultraslow, minute-long afterhyperpolarization (usAHP) in network neurons following locomotor episodes. The usAHP is mediated by an activity- and sodium spike-dependent enhancement of electrogenic Na(+)/K(+) pump function. By integrating spike frequency over time and linking the membrane potential of spinal neurons to network performance, the usAHP plays a dynamic role in short-term motor memory. Because Na(+)/K(+) pumps are ubiquitously expressed in neurons of all animals and because sodium spikes inevitably accompany network activity, the usAHP may represent a phylogenetically conserved but largely overlooked mechanism for short-term memory of neural network function.     

Using System Emulation to Model Next-Generation Shared Virtual Memory Clusters

Recently much effort has been spent on providing a shared address space abstraction on clusters of small-scale symmetric multiprocessors. However, advances in technology will soon make it possible to construct these clusters with larger-scale cc-NUMA nodes, connected with non-coherent networks that offer latencies and bandwidth comparable to interconnection networks used in hardware cache-coherent systems. The shared memory abstraction can be provided on these systems in software across nodes and hardware within nodes.Recent simulation results have demonstrated that certain features of modern system area networks can be used to greatly reduce shared virtual memory (SVM) overheads [5,19]. In this work we leverage these results and we use detailed system emulation to investigate building future software shared memory clusters. We use an existing, large-scale hardware cache-coherent system with 64 processors to emulate a complete future cluster. We port our existing infrastructure (communication layer and shared memory protocol) on this system and study the behavior of a set of real applications. We present results for both 32- and 64-processor system configurations.We find that: (i) System emulation is invaluable in quantifying potential benefits from changes in the technology of commodity components. More importantly, it reveals potential problems in future systems that are easily overlooked in simulation studies. Thus, system emulation should be used along with other modeling techniques (e.g., simulation, implementation) to investigate future trends. (ii) Our work shows that current SVM protocols can only partially take advantage of faster interconnects and wider nodes due to operating system and architectural implications. We quantify the related issues and identify the areas where more research is required for future SVM clusters.     

体外培养的神经元网络可塑性

神经元网络是大脑执行高级认知行为的结构基础,研究证明学习记忆及神经退行性疾病与神经元网络可塑性密切相关。因此,揭示调控和改变神经元网络可塑性的机制对理解神经系统信息交互以及疾病治疗具有重大意义。目前,基于微电极阵列（microelectrode array, MEA）培养的神经元网络是体外探究学习和记忆机制的理想模型,同时针对该模型的研究为预防和治疗神经退行性疾病提供了独特的视角。本文综述了基于MEA采集体外培养神经元网络的放电信号来构建功能网络的相关研究,分别从二维神经元网络和三维脑类器官发育,以及开环和闭环电刺激对神经元网络可塑性影响的角度,总结了体外培养神经元网络可塑性的相关研究,最后对该方向的应用前景进行了展望。     

Reliable prediction of T-cell epitopes using neural networks with novel sequence representations

In this paper we describe an improved neural network method to predict T-cell class I epitopes. A novel input representation has been developed consisting of a combination of sparse encoding, Blosum encoding, and input derived from hidden Markov models. We demonstrate that the combination of several neural networks derived using different sequence-encoding schemes has a performance superior to neural networks derived using a single sequence-encoding scheme. The new method is shown to have a performance that is substantially higher than that of other methods. By use of mutual information calculations we show that peptides that bind to the HLA A*0204 complex display signal of higher order sequence correlations. Neural networks are ideally suited to integrate such higher order correlations when predicting the binding affinity. It is this feature combined with the use of several neural networks derived from different and novel sequence-encoding schemes and the ability of the neural network to be trained on data consisting of continuous binding affinities that gives the new method an improved performance. The difference in predictive performance between the neural network methods and that of the matrix-driven methods is found to be most significant for peptides that bind strongly to the HLA molecule, confirming that the signal of higher order sequence correlation is most strongly present in high-binding peptides. Finally, we use the method to predict T-cell epitopes for the genome of hepatitis C virus and discuss possible applications of the prediction method to guide the process of rational vaccine design.     

XenLoop: a transparent high performance inter-VM network loopback

Advances in virtualization technology have focused mainly on strengthening the isolation barrier between virtual machines (VMs) that are co-resident within a single physical machine. At the same time, a large category of communication intensive distributed applications and software components exist, such as web services, high performance grid applications, transaction processing, and graphics rendering, that often wish to communicate across this isolation barrier with other endpoints on co-resident VMs. State of the art inter-VM communication mechanisms do not adequately address the requirements of such applications. TCP/UDP based network communication tends to perform poorly when used between co-resident VMs, but has the advantage of being transparent to user applications. Other solutions exploit inter-domain shared memory mechanisms to improve communication latency and bandwidth, but require applications or user libraries to be rewritten against customized APIs—something not practical for a large majority of distributed applications. In this paper, we present the design and implementation of a fully transparent and high performance inter-VM network loopback channel, called XenLoop, in the Xen virtual machine environment. XenLoop does not sacrifice user-level transparency and yet achieves high communication performance between co-resident guest VMs. XenLoop intercepts outgoing network packets beneath the network layer and shepherds the packets destined to co-resident VMs through a high-speed inter-VM shared memory channel that bypasses the virtualized network interface. Guest VMs using XenLoop can migrate transparently across machines without disrupting ongoing network communications, and seamlessly switch between the standard network path and the XenLoop channel. In our evaluation using a number of unmodified benchmarks, we observe that XenLoop can reduce the inter-VM round trip latency by up to a factor of 5 and increase bandwidth by a up to a factor of 6.