• KIPS Transactions on Computer and Communication Systems
• /
• v.10 no.5
• /
• pp.123-136
• /
• 2021

RPC (Remote Procedure Call)-based Graphics Processing Unit (GPU) virtualization technology is one of the technologies for sharing GPUs with multiple user virtual machines. However, in a cloud environment, unlike CPU or memory, general GPUs do not provide a resource isolation technology that can limit the resource usage of virtual machines. In particular, in an RPC-based virtualization environment, since GPU tasks executed in each virtual machine are performed in the form of multi-process, the lack of resource isolation technology causes performance degradation due to resource competition. In addition, the GPU memory competition accelerates the performance degradation as the resource demand of the virtual machines increases, and the fairness decreases because it cannot guarantee equal performance between virtual machines. This paper, in the RPC-based GPU virtualization environment, analyzes the performance degradation problem caused by resource contention when the GPU memory requirement of virtual machines exceeds the available GPU memory capacity and proposes a GPU memory management technique to solve this problem. Also, experiments show that the GPU memory management technique proposed in this paper can improve the performance of GPGPU tasks.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.5
• /
• pp.11-19
• /
• 2022

In this paper, we propose a fault tolerant embedded system using memory redundancy on the full-virtualization based virtual machine monitor. The proposed virtual machine monitor first virtualizes main memory of embedded system utilizing efficient shadow page table scheme so that the embedded system runs as a virtual machine on the virtual machine monitor. The virtual machine monitor makes the backup of the embedded system run as another virtual machine by copying memory contents of the embedded system into memory space of backup system according to predefined schedules. When an error occurs in the target virtual machine, the corresponding standby virtual machine takes the role of target virtual machine and continues its operation. Performance evaluation studies show that such backups and switches of virtual machines are performed with minor performance degradation.

• Journal of KIISE:Computer Systems and Theory
• /
• v.37 no.3
• /
• pp.147-160
• /
• 2010

The system virtualization shows interest in the consolidation of servers for the efficient utilization of system resources. There are many various researches to utilize a server machine more efficiently through the system virtualization technique, and improve performance of the virtualization software. These researches have studied with the activity to control the resource allocation of virtual machines dynamically focused on CPU, or to manage resources in the cross-machine using the migration. However, the researches of the memory management have been wholly lacking. In this respect, the use of memory is limited to allocate the memory statically to virtual machine in server consolidation. Unfortunately, the static allocation of the memory causes a great quantity of the idle memory and decreases the memory utilization. The underutilization of the memory makes other side effects such as the load of other system resources or the performance degradation of services in virtual machines. In this paper, we suggest the dynamic allocation of the memory in Xen to control the memory allocation of virtual machines for the utilization without the performance degradation. Using AR model for the prediction of the memory usage and ACO (Ant Colony Optimization) algorithm for optimizing the memory utilization, the system operates more virtual machines without the performance degradation of servers. Accordingly, we have obtained 1.4 times better utilization than the static allocation.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.11
• /
• pp.5234-5251
• /
• 2018

It is well-known that virtualization technology can bring many benefits not only to users but also to service providers. From the view of system security and resource utility, higher resource sharing degree and higher system reliability can be obtained by the introduction of virtualization technology in distributed cloud. The small size time-sharing multiplexing technology which is based on virtual machine in distributed cloud platform can enhance the resource utilization effectively by server consolidation. In this paper, the concept of memory block and user satisfaction is redefined combined with user requirements. According to the unbalanced memory resource states and user preference requirements in multi-virtual machine environments, a model of proper memory resource allocation is proposed combined with memory block and user satisfaction, and at the same time a memory optimization allocation algorithm is proposed which is based on virtual memory block, makespan and user satisfaction under the premise of an orderly physical nodes states also. In the algorithm, a memory optimal problem can be transformed into a resource workload balance problem. All the virtual machine tasks are simulated in Cloudsim platform. And the experimental results show that the problem of virtual machine memory resource allocation can be solved flexibly and efficiently.

• Journal of KIISE
• /
• v.44 no.10
• /
• pp.1005-1018
• /
• 2017

Recently, virtualization has become an essential component of cloud computing due to its various strengths, including maximizing server resource utilization, easy-to-maintain software, and enhanced data protection. However, since virtualization allows sharing physical resources among the VMs, the system performance can be deteriorated due to device contentions. In this paper, we first investigate the I/O overhead based on the number of VMs on the same server platform and analyze the block I/O process of the KVM hypervisor. We also propose an in-memory block cache mechanism, called QBic, to overcome I/O virtualization latency. QBic is capable of monitoring the block I/O process of the hypervisor and stores the data with a high access frequency in the cache. As a result, QBic provides a fast response for VMs and reduces the I/O contention to physical devices. Finally, we present a performance measurement of QBic to verify its effectiveness.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.12
• /
• pp.5375-5400
• /
• 2016

IOMMU is a hardware unit that is indispensable for DMA. Besides address translation and remapping, it also provides I/O virtual address space isolation among devices and memory access control on DMA transactions. However, currently commodity virtualization platforms lack of IOMMU virtualization, so that the virtual machines are vulnerable to DMA security threats. Previous works focus only on DMA security problem of directly assigned devices. Moreover, these solutions either introduce significant overhead or require modifications on the guest OS to optimize performance, and none can achieve high I/O efficiency and good compatibility with the guest OS simultaneously, which are both necessary for production environments. However, for simulated virtual devices the DMA security problem also exists, and previous works cannot solve this problem. The reason behind that is IOMMU circuits on the host do not work for this kind of devices as DMA operations of which are simulated by memory copy of CPU. Motivated by the above observations, we propose an IOMMU para-virtualization solution called PVIOMMU, which provides general functionalities especially DMA security guarantees for both directly assigned devices and simulated devices. The prototype of PVIOMMU is implemented in Qemu/KVM based on the virtio framework and can be dynamically loaded into guest kernel as a module, As a result, modifying and rebuilding guest kernel are not required. In addition, the device model of Qemu is revised to implement DMA access control by separating the device simulator from the address space of the guest virtual machine. Experimental evaluations on three kinds of network devices including Intel I210 (1Gbps), simulated E1000 (1Gbps) and IB ConnectX-3 (40Gbps) show that, PVIOMMU introduces little overhead on DMA transactions, and in general the network I/O performance is close to that in the native KVM implementation without IOMMU virtualization.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.8 no.11
• /
• pp.1741-1748
• /
• 2013

In this paper, we proposed new virtualization technology that is more convenient and stable in local computing environment, then found technique elements need to desktop virtualization which is based on clients in various virtualization technologies. After running excution of process explorer utility in user level virtualization and VMWare, we found memory capacity that is used 30.1MB in VMWare and 16.6MB in user level virtualization respectively to compare private bytes each of process. We found no significant difference of CPU utilization which is executed application program in local computing environment and user domain with user level virtualization. In this result, proposed virtualization technology is able to minimize performance degradation of local computing environment.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.8 no.3
• /
• pp.145-153
• /
• 2013

Virtualization technology has been applied in IA-32 based server or desktop systems. Recently it has been applied in ARM based mobile systems. Virtualization technology provides many useful features that are not possible in operating system level such as isolation, interposition, encapsulation and portability. In this research, we implement an ARM based VMM(Virtual Machine Monitor) by using the following techniques. First, we use "emulation" to virtualize the processor. Second, we use "shadow page tables" to virtualize the memory. Finally, we use a simple "pass-through I/O" to virtualize the device. Currently the VMM runs ARM Linux kernel 3.4.4 on a BeagleBoard-xM, and we evaluated the performance of the VMM using lmbench and dhrystone. The result of the evaluation shows that our VMM is slower than Xen on ARM that is implemented using paravirtualization but has good performance among the VMMs using full-virtualization.

• KIISE Transactions on Computing Practices
• /
• v.21 no.12
• /
• pp.792-797
• /
• 2015

Effective profiling diagnoses the failure of the system and informs risk. If a failure in the target system occurs, it is impossible to diagnose more than one of the exiting tools. In this respect, monitoring of the system based on virtualization is useful. We present in this paper a monitoring framework that uses the characteristics of hardware virtualization to prevent side-effects from a target guest, and uses dynamic binary instrumentation with instruction-level trapping based on hardware virtualization to achieve efficiency and flexibility. We also present examples of some applications that use this framework. The framework provides tracing of guest kernel function, memory dump, and debugging that uses GDB stub with GDB remote protocol. The experimental evaluation of our prototype shows that the monitoring framework incurs at most 2% write memory performance overhead for end users.

• The KIPS Transactions:PartC
• /
• v.18C no.3
• /
• pp.157-166
• /
• 2011

The virtualization layer is executed in higher authority layer than kernel layer and suitable for monitoring operating systems. However, existing virtualization monitoring systems provide simple information about the usage rate of CPU or memory. In this paper, the monitoring system using full virtualization technique is proposed, which can monitor virtual machine's dynamic kernel object as memory, register, GDT, IDT and system call table. To verify the monitoring system, the proposed system was implemented based on KVM(Kernel-based Virtual Machine) with full virtualization that is directly applied to linux kernel without any modification. The proposed system consists of KvmAccess module to access KVM's internal object and API to provide other external modules with monitoring result. In experiments, the CPU utilization for monitoring operations in the proposed monitering system is 0.35% when the system is monitored with 1-second period. The proposed monitoring system has a little performance degradation.