Hardness of Approximation and Greedy Algorithms for the Adaptation Problem in Virtual Environments

As discovered in our previous benchmark works, a small number of workloads in PARSEC benchmark suite suffer from heavy performance loss in a virtual execution environment, of which the major loss exhibits fairly a strong connection with the thread synchronization operations. This paper examines one workload of this kind that makes heavy use of thread synchronization operations, and shows the performance impact by futex system call for applications in a virtual execution environment. By measuring the occurrence of the subroutines running at particular time costs, the most timing-decisive parts of the application in a virtual execution environment are located. Proper thread scheduling measure for a virtual execution environment are also proposed.

Recently, automatic judgment systems have been widely used in various computer science lectures, and these systems are mainly developed for languages with a console‐based practice environment. On the other hand, embedded systems such as Arduino are executed in a hardware‐based environment and their verification is conducted by manually checking the operation of the hardware. Therefore, it is not easy to automatically judge the Arduino system developed by students. In the embedded system, hardware and software have to be developed at the same time, and it increases effort and time for the instructors to teach students, especially in practice‐based learning. The study in this article proposes an Arduino practice judgment system based on the function execution log in virtual execution environment. It performs both Fritzing‐based hardware configuration checking and source code testing based on virtual execution environment, where hardware operations are replaced by mock‐up functions. In addition, more diverse practices are possible by providing an experimental environment where students can freely enter input data and check the results. A case study shows that the developed system can be applied to the lectures and help reduce instructors' work and increase students' understanding through the system.

During recent years, microprocessor energy consumption has been surging and efforts to reduce power and energy have received a lot of attention. At the same time, virtual execution environments (VEEs), such as Java virtual machines, have grown in popularity. Hence, it is important to evaluate the impact of virtual execution environments on microprocessor energy consumption. This paper characterizes the energy and power impact of two important components of VEEs, Just-in-time(JIT) optimization and garbage collection. We find that by reducing instruction counts, JIT optimization significantly reduces energy consumption, while garbage collection incurs runtime overhead that consumes more energy. Importantly, both JIT optimization and garbage collection decrease the average power dissipated by a program. Detailed analysis reveals that both JIT optimizer and JIT optimized code dissipate less power than un-optimized code. On the other hand, being memory bound and with low ILP, the garbage collector dissipates less power than the application code, but rarely affects the average power of the latter.Adaptive microarchitectures are another recent trend for energy reduction where microarchitectural resources can be dynamically tuned to match program runtime requirements. This research reveals that both JIT optimization and garbage collection alter a program's behavior and runtime requirements, which considerably affects the adaptation of configurable hardware units, and influences the overall energy consumption. This work also demonstrates that the adaptation preferences of the two VEE services differ substantially from those of the application code. Both VEE services prefer a simple core for high energy reduction. On the other hand, the JIT optimizer usually requires larger data caches, while the garbage collector rarely benefits from large data caches. The insights gained in this paper point to novel techniques that can further reduce microprocessor energy consumption.

The security of hardware-software systems is at risk from a wide range of attack vectors that appear at various stages during the execution of machine code. The existing approaches for repairing software defects have numerous restrictions with respect to their applicability and functionality and to the range of vulnerabilities that can be identified and eliminated.We propose an approach for removing software errors in program code that is based on just-in-time compilation in a virtual execution environment. The virtual environment uses static, dynamic, and hybrid analyses of the intermediate representation of vulnerable code and re-compiles such code to be safe.The language of code annotations allows us to manage static and dynamic analyses and code transformations. We can change the level of analysis and the amount of time spent on such analyses by dynamically adapting the precision.

It is our pleasure to welcome you to the 6th ACM SIGPLAN/SIGOPS Conference on Virtual Execution Environments (VEE'10). As the leading conference for presentation of research results on all aspects of virtualization, VEE brings together researchers representing a diverse set of interests. As with previous VEEs, this year's program represents an exciting mix of papers ranging from hardware virtualization to virtual machines for programming languages. In selecting papers, the program committee placed a high priority on choosing work that would be broadly interesting and applicable. This year's conference has upheld a number of other traditions: as in the previous three years, the conference was co-chaired by researchers from both the operating systems and programming languages communities. The program committee also represents a mix of research backgrounds, and features strong representation from industrial research organizations. VEE is again co-located with the International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS). VEE'10 had 50 submissions, 15 of which were chosen for presentation at the conference. The accepted papers span a wide range of virtualization, broadly construed, and we are confident they will make for an interesting conference.

A virtual machine is a software replica of an underlying real machine. Multiple virtual machines can operate on the same host machine concurrently, without interfere each other. Such concept is becoming valuable in production computing systems, due to its benefits in terms of costs and portability. As they provide a strong isolation between the virtual environment and the underlying real system, virtual machines can also be used to improve the security of a computer system in face of attacks to its network services. This work presents a new approach to achieve this goal, by applying intrusion detection techniques to virtual machine based systems, thus keeping the intrusion detection system out of reach from intruders. The results obtained from a prototype implementation confirm the usefulness of this approach.

Performance assessment of 40 Gbit/s off-the-shelf network cards for virtual network probes in 5G networks

In this paper, an innovative approach to controlling temperature was made. Using Unity as our chosen platform, we crafted a virtual environment, that replicates the real-world temperature change dynamics. For this, a virtual environment mimics an enclosed area with a virtual heat source. Additionally, this environment’s area and air density were also taken into consideration. Then, a virtual sensor and cooler were added for detecting and adjusting the environment temperature. Finally, the ML agent package was integrated for controlling temperature with RL. This integration consisted of adding a script that inherits from the agent class and has environment, sensor, and cooler as variables. In each epi-sode, there was a time duration in which both environment's temperature change was calculated and the reward was set to be given only when the environment's temperature was at a certain range. The empirical findings show the efficiency of our approach, show-casing the system's ability to adapt and optimize temperature regulation parameters over successive iterations. By successfully integrating Unity and ML agents, our study under-scores the transformative potential of harnessing virtual environments and machine learning for advancing temperature control technologies. Keywords: unity, ml agents, temperature control, virtual environment, reinforcement learning.

Network forensic investigation in OpenFlow networks with ForCon

Nowadays, Network Address Translation (NAT) is widely used to allow multiple devices within a private network to make use of a less number of public IP addresses. NAT Overloading or Port Address Translation (PAT) is an extension of NAT that can translate both the IP address and the port number of a packet in order to identify which inside local address each packet belongs to. PAT is often used in a virtual environment, where multiple virtual machines are connected to the Internet by using the host machine's IP address. However, an apparent downside of PAT is the fact that when all of the ports are used, no more new outbound connection could be made from the local addresses. In this paper, we present Slow-port-exhaustion DoS Attack on a virtual network, a new type of DoS Attack that exploits some flaws of the TCP protocol. In this attack, a compromised internal virtual machine with a low amount of attack bandwidth can occupy host machine's ports for a long time and therefore makes other machines on the same virtual network could not connect to the external network. We created a virtual network with PAT implemented gateway and perform the experimental attack. In the analysis, we explore a gateway's behavior that could benefit this kind of attack. We also introduce some countermeasures against this kind of attack.

The standard tools for cloud computing — processor and network virtualization — make it difficult to achieve dependability, both in terms of real time operations and fault tolerance. Virtualization multiplexes virtual resources onto physical ones, typically by time division or statistical multiplexing. Time, in the virtual machine, is therefore as virtual as the machine itself. And fault tolerance is difficult to achieve when redundancy and independent failure in the virtual environment do not necessarily map to those properties in the physical environment. Virtualization adds a level of indirection that creates overhead, and makes it all but impossible to achieve predictable performance. Osprey uses an alternative to virtualization that achieves the same goals of scalability and flexibility but carries neither the overhead of virtualization, nor the restrictions on dependability. The result is a programming environment that achieves most of the compatibility offered by traditional virtualization efforts and provides much better and much more predictable performance. One technique we use is called Library OS, which stems from high-performance computing. The technique consists of linking applications with a library that implements most services normally provided by the operating system, creating an application that can run practically stand alone, or at least with a very minimal operating system. The Library OS approach moves the boundary between application and operating system down to a level where interactions with the operating system consist of sending/receiving messages (e.g., network packets) and scheduling resources (processor, memory, network bandwidth, and device access). These interactions, as we demonstrate, form a relatively weak bond between an application and the particular instance of the operating system on which it runs — one that can be broken and re-established elsewhere. In fact, we make sure this is the case. Legacy applications that cannot be recompiled or relinked can make use of a Library OS server that runs as a tandem process along with the legacy application processes. System calls from the legacy process are catapulted into the Library OS server which executes them. Applications can still migrate, taking their server process along with them. © 2012 Alcatel-Lucent.

Unlike physical systems consisting of hardware, virtual machines are conceptual computers in logical configurations made of software. Although programs implemented in virtual machine environments have platform-independent characteristics, their performance is low in terms of implementation speed and they cannot implement those parts that are dependent on the platform by themselves. In addition, they have the shortcoming of being unable to reuse those libraries and programs that are written in the native language. In the present paper, a study for provision of native interfaces of smart virtual machines (SVMs) is introduced. The native interfaces of existing systems were analyzed to define a native interface protocol for SVMs, an interface definition language and an interface generator were designed for effective use of the defined native interface protocol, and the operation of the implemented native interface was verified through experiments. The native interface proposed in the present paper consists of items and rules necessary for information transmission between virtual machine environments and native environments, interface definition files necessary to deliver information on native functions to virtual machines, and functions and data types necessary for native environment to receive information from virtual machine environments. Through such native interfaces, virtual machine environment developers can use environment-dependent function without modifying virtual machine environments.

This paper shows the early research findings of utilizing a virtual reality environment as an educational tool for the operation of a computerized numerical control (CNC) milling machine. Based off of previous work, the Advanced Virtual Machining Lab (AVML), this project features an environment in which a virtual CNC machine is fully operable, designed to allow STEM students and training professionals to learn the use of the CNC machine without the need to be in a physical lab. Users operate in the virtual environment using an immersive virtual reality headset (i.e. Oculus Rift) and standard input devices (i.e. mouse and keyboard), both of which combined make for easy movement and realistic visuals. On-screen tutorials allow users to learn about what they need to do to operate the machine without the need for outside instruction. While designing and perfecting this environment has been the primary focus of this project thus far, the research goal is to test the ease of use and the pedagogical effectiveness of the immersive technology as it relates to education in STEM fields. Initial usability studies for this environment featured students from a CAD/CAM-Theory and Advanced Applications (ME 54600) course at a Midwestern urban institution. Results from the study were tabulated with a survey using a four-point Likert scale and several open-ended questions. Findings from the survey indicated that the majority of users found the environment realistic and easy to navigate, in addition to finding the immersive technology to be beneficial. Many also indicated that they felt comfortable navigating the environment without the need for additional assistance from the survey proctors. Full details on the usability study, including data and discussion, can be found in this paper. The general consensus from the study was that, while some features needed refinement, the immersive environment helped them learn about the operation of a CNC machine. An additional comparative study will be undertaken to evaluate pedagogical effectiveness.

Hadoop MapReduce as a service from cloud is widely used by various research, and commercial communities. Hadoop MapReduce is typically offered as a service hosted on virtualized environment in Cloud Data-Center. Cluster of virtual machines for MapReduce is placed across racks in Cloud Data-Center to achieve fault tolerance. But, it negatively introduces dynamic/heterogeneous performance for virtual machines due to hardware heterogeneity and co-located virtual machine's interference, which cause varying latency for same task. Alongside, curbing number of intermediate records and placing reduce tasks on right virtual node are also important to minimize MapReduce job latency further. In this paper, we introduce Multi-Level Per Node Combiner to minimize the number of intermediate records and Dynamic Ranking based MapReduce Job Scheduler to place reduce tasks on right virtual machine to minimize MapReduce job latency by exploiting dynamic performance of virtual machines. To experiment and evaluate, we launched 29 virtual machines hosted in eight different physical machines to run wordcount job on PUMA dataset. Our proposed methodology improves overall job latency up to 33% for wordcount job.

We have developed a middleware system that integrates and extends virtual machine and virtual network technologies to support mutually isolated virtual distributed environments in shared infrastructures like the grid and the PlanetLab overlay infrastructure. Integrating virtual network and on-demand virtual machine creation and customization technologies makes virtual distributed environments a reality. The Violin-based middleware system integrates and enhances such technologies to create virtual distributed environments.