The Experts below are selected from a list of 90819 Experts worldwide ranked by ideXlab platform

Priya Chandran - One of the best experts on this subject based on the ideXlab platform.

  • dynamic partitioning of Physical Memory among virtual machines asmi architectural support for Memory isolation
    ACM Symposium on Applied Computing, 2016
    Co-Authors: R Jithin, Priya Chandran
    Abstract:

    It is an open challenge for virtualization technology architects to provide security to Virtual Machine (VM), in the presence of an infected hypervisor, without much compromise on performance. A few hardware modifications have been introduced by manufactures like Intel and AMD to provide a secure VM environment with low performance degradation. These solutions are unable to provide VM isolation in the presence of an infected hypervisor. In this paper we propose a novel Memory architecture model, that can achieve a secure Physical Memory region to each VM without performance degradation.

  • dynamic partitioning of Physical Memory among virtual machines asmi architectural support for Memory isolation
    arXiv: Hardware Architecture, 2015
    Co-Authors: R Jithin, Priya Chandran
    Abstract:

    Cloud computing relies on secure and efficient virtualization. Software level security solutions compromise the performance of virtual machines (VMs), as a large amount of computational power would be utilized for running the security modules. Moreover, software solutions are only as secure as the level that they work on. For example a security module on a hypervisor cannot provide security in the presence of an infected hypervisor. It is a challenge for virtualization technology architects to enhance the security of VMs without degrading their performance. Currently available server machines are not fully equipped to support a secure VM environment without compromising on performance. A few hardware modifications have been introduced by manufactures like Intel and AMD to provide a secure VM environment with low performance degradation. In this paper we propose a novel Memory architecture model named \textit{ Architectural Support for Memory Isolation(ASMI)}, that can achieve a true isolated Physical Memory region to each VM without degrading performance. Along with true Memory isolation, ASMI is designed to provide lower Memory access times, better utilization of available Memory, support for DMA isolation and support for platform independence for users of VMs.

Alexander Arth - One of the best experts on this subject based on the ideXlab platform.

  • Rambrain - a library for virtually extending Physical Memory
    SoftwareX, 2017
    Co-Authors: Maximilian Imgrund, Alexander Arth
    Abstract:

    Abstract We introduce Rambrain, a user space C++ library that manages Memory consumption of data-intense applications. Using Rambrain, one can overcommit Memory beyond the size of Physical Memory present in the system. While there exist other more advanced techniques to solve this problem, Rambrain focuses on saving development time by providing a fast, general and easy-to-use solution. Rambrain takes care of temporarily swapping out data to disk and can handle multiples of the Physical Memory size present. Rambrain is thread-safe, OpenMP and MPI compatible and supports asynchronous I/O. The library is designed to require minimal changes to existing programs and pose only a small overhead.

  • Rambrain - a library for virtually extending Physical Memory
    arXiv: Distributed Parallel and Cluster Computing, 2015
    Co-Authors: Maximilian Imgrund, Alexander Arth
    Abstract:

    We introduce Rambrain, a user space library that manages Memory consumption of your code. Using Rambrain you can overcommit Memory over the size of Physical Memory present in the system. Rambrain takes care of temporarily swapping out data to disk and can handle multiples of the Physical Memory size present. Rambrain is thread-safe, OpenMP and MPI compatible and supports Asynchronous IO. The library was designed to require minimal changes to existing programs and to be easy to use.

R Jithin - One of the best experts on this subject based on the ideXlab platform.

  • dynamic partitioning of Physical Memory among virtual machines asmi architectural support for Memory isolation
    ACM Symposium on Applied Computing, 2016
    Co-Authors: R Jithin, Priya Chandran
    Abstract:

    It is an open challenge for virtualization technology architects to provide security to Virtual Machine (VM), in the presence of an infected hypervisor, without much compromise on performance. A few hardware modifications have been introduced by manufactures like Intel and AMD to provide a secure VM environment with low performance degradation. These solutions are unable to provide VM isolation in the presence of an infected hypervisor. In this paper we propose a novel Memory architecture model, that can achieve a secure Physical Memory region to each VM without performance degradation.

  • dynamic partitioning of Physical Memory among virtual machines asmi architectural support for Memory isolation
    arXiv: Hardware Architecture, 2015
    Co-Authors: R Jithin, Priya Chandran
    Abstract:

    Cloud computing relies on secure and efficient virtualization. Software level security solutions compromise the performance of virtual machines (VMs), as a large amount of computational power would be utilized for running the security modules. Moreover, software solutions are only as secure as the level that they work on. For example a security module on a hypervisor cannot provide security in the presence of an infected hypervisor. It is a challenge for virtualization technology architects to enhance the security of VMs without degrading their performance. Currently available server machines are not fully equipped to support a secure VM environment without compromising on performance. A few hardware modifications have been introduced by manufactures like Intel and AMD to provide a secure VM environment with low performance degradation. In this paper we propose a novel Memory architecture model named \textit{ Architectural Support for Memory Isolation(ASMI)}, that can achieve a true isolated Physical Memory region to each VM without degrading performance. Along with true Memory isolation, ASMI is designed to provide lower Memory access times, better utilization of available Memory, support for DMA isolation and support for platform independence for users of VMs.

Jill Slay - One of the best experts on this subject based on the ideXlab platform.

  • ARES - Recovery of Pidgin Chat Communication Artefacts from Physical Memory: A Pilot Test to Determine Feasibility
    2011 Sixth International Conference on Availability Reliability and Security, 2011
    Co-Authors: Matthew Simon, Jill Slay
    Abstract:

    This research describes a study that looks at the feasibility of extracting remnant information about an instant message client from Physical Memory. The research goal was to gather information about the target application in order to assess the viability of creating methods to recover specific data about its use. The study consists of a formal experiment where the application is used and the Physical Memory collected at various points. The Memory image was then interrogated to assess whether remnant data could be recovered. The study shows that it is feasible to recover data about the target application.

  • recovery of skype application activity data from Physical Memory
    Availability Reliability and Security, 2010
    Co-Authors: Matthew Simon, Jill Slay
    Abstract:

    The use of Internet based communication technologies has become more prevalent in recent years. Technologies such as Skype provide a highly secure and decentralised method of communication. These technologies may also leave little evidence on static media causing conventional digital forensic processes to be ineffective. This research looks at exploiting Physical Memory to recover evidence from Internet based communication technologies where conventional methods cannot. The paper first proposes a set of generic target artefacts that defines information that may be targeted for recovery and the meaning that can be inferred from this. A controlled test was then undertaken where Skype was executed and the Memory from the target machine collected. The analysis showed that it is feasible to recover the target data as applied to Skype, which would not be otherwise available. As this is the first set of tests of a series, the future direction is also discussed.

  • ARES - Recovery of Skype Application Activity Data from Physical Memory
    2010 International Conference on Availability Reliability and Security, 2010
    Co-Authors: Matthew Simon, Jill Slay
    Abstract:

    The use of Internet based communication technologies has become more prevalent in recent years. Technologies such as Skype provide a highly secure and decentralised method of communication. These technologies may also leave little evidence on static media causing conventional digital forensic processes to be ineffective. This research looks at exploiting Physical Memory to recover evidence from Internet based communication technologies where conventional methods cannot. The paper first proposes a set of generic target artefacts that defines information that may be targeted for recovery and the meaning that can be inferred from this. A controlled test was then undertaken where Skype was executed and the Memory from the target machine collected. The analysis showed that it is feasible to recover the target data as applied to Skype, which would not be otherwise available. As this is the first set of tests of a series, the future direction is also discussed.

  • What are you Looking for: Identification of Remnant Communication Artefacts in Physical Memory
    2010
    Co-Authors: Matthew Simon, Jill Slay
    Abstract:

    Law enforcement has sound methods for investigating and obtaining data about targets that are using traditional communication services such as the Public Switched Telephone Network. The Internet as a data transfer medium is a vastly different paradigm to that of traditional telephony networks. Information about targets using Internet communication technologies cannot be obtained using the same methods used for traditional communication. There has been an identified need for methods to obtain information on targets that have been using Internet communication methods. The acquisition and analysis of Physical Memory has been proposed as a vector for the recovery of such information. In order to investigate Memory analysis and communication technologies, it is necessary to define the types of data that investigators should look for. To this end, the concept of a set of data artefacts has been defined that contains generic data types that are inherent to all Internet based communication applications. To demonstrate the utility of the concept, a case study is presented that applies the artefacts to Skype.

Rami Melhem - One of the best experts on this subject based on the ideXlab platform.

  • supporting superpages in non contiguous Physical Memory
    High-Performance Computer Architecture, 2015
    Co-Authors: Miao Zhou, Bruce R Childers, Daniel Mosse, Rami Melhem
    Abstract:

    For Memory-intensiv e workloads with large Memory footprints, superpages are effective to avoid address translation overhead, which can be a critical performance bottleneck. A superpage is a large virtual Memory page that is mapped to an equivalently-sized amount of contiguous Physical Memory pages. Superpage mapping assumes Physical Memory does not contain retired pages, which is an important technique to improve Memory resilience: the OS avoids allocating Physical pages that have detected errors. Retired pages create unusable "holes" in the Physical Memory. We show that even a small percentage of retired pages makes it very difficult to find enough contiguous Memory to form superpages. To address this problem, we propose GTSM, or gap-tolerant sequential mapping, that allows superpages to be formed even in the presence of retired Physical pages. A new page table format is also proposed to support GTSM. This format has similar storage efficiency as traditional superpaging to hold address translations in the last-level cache. To further compress the page table and improve cache hit rates for address translation in large Memory footprint workloads, we also propose an extended format that reduces the page table size by 50%. In comparison to an ideal Memory without any retired Physical pages, we show that our technique, with retired pages, achieves nearly 96.8% of the performance of traditional 2MB superpaging.

  • HPCA - Supporting superpages in non-contiguous Physical Memory
    2015 IEEE 21st International Symposium on High Performance Computer Architecture (HPCA), 2015
    Co-Authors: Miao Zhou, Bruce R Childers, Daniel Mosse, Rami Melhem
    Abstract:

    For Memory-intensiv e workloads with large Memory footprints, superpages are effective to avoid address translation overhead, which can be a critical performance bottleneck. A superpage is a large virtual Memory page that is mapped to an equivalently-sized amount of contiguous Physical Memory pages. Superpage mapping assumes Physical Memory does not contain retired pages, which is an important technique to improve Memory resilience: the OS avoids allocating Physical pages that have detected errors. Retired pages create unusable "holes" in the Physical Memory. We show that even a small percentage of retired pages makes it very difficult to find enough contiguous Memory to form superpages. To address this problem, we propose GTSM, or gap-tolerant sequential mapping, that allows superpages to be formed even in the presence of retired Physical pages. A new page table format is also proposed to support GTSM. This format has similar storage efficiency as traditional superpaging to hold address translations in the last-level cache. To further compress the page table and improve cache hit rates for address translation in large Memory footprint workloads, we also propose an extended format that reduces the page table size by 50%. In comparison to an ideal Memory without any retired Physical pages, we show that our technique, with retired pages, achieves nearly 96.8% of the performance of traditional 2MB superpaging.