SIMDOM: A framework for SIMD instruction translation and offloading in heterogeneous mobile architectures

Shuja, Junaid; Gani, Abdullah; Ko, Kwangman; So, Kyoung-Young; Mustafa, Saad; Madani, Sajjad A.; Khan, Muhammad Khurram

doi:10.1002/ett.3174

Cited by 27 publications

(18 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another framework called SIMDOM was presented for single instruction, multiple data (SIMD) instruction translation and offloading for mobile devices in cloud and edge environments. The SIMDOM framework reduced the execution overhead of migrated vectorized multimedia application by using vector‐to‐vector instruction mappings.…”

Section: Related Workmentioning

confidence: 99%

EdgePlace: Availability‐aware placement for chained mobile edge applications

Zhu

Huang

2018

Trans Emerging Tel Tech

View full text Add to dashboard Cite

Mobile edge computing (MEC) literally pushes cloud computing from remote datacenters to the life radius of end users. By leveraging the widely adopted European Telecommunications Standards Institute network function virtualization architecture, MEC provisions elastic and resilient mobile edge applications with proximity. Typical MEC virtualization infrastructure allows a configurable placement policy to deploy mobile edge applications as virtual machines (VMs): affinity can be used to put VMs on the same host for inter‐VM networking performance, whereas anti‐affinity is to separate VMs for high availability. In this paper, we propose a novel model to track the availability and cost impact from placement policy changes of the mobile edge applications. We formulate our model as a stochastic programming problem. To minimize the complexity challenge, we also propose a heuristic algorithm called EdgePlace. With our model, the unit resource cost increases when there are less resources left on a host. Applying affinity would take up more resources of the host but saves network bandwidth cost because of colocation. When enforcing anti‐affinity, experimental results show increases of both availability and interhost network bandwidth cost. For applications with different resource requirements, our model is able to find their sweet points with the consideration of both resource cost and application availability, which is vital in a less robust MEC environment.

show abstract

Section: Related Workmentioning

confidence: 99%

EdgePlace: Availability‐aware placement for chained mobile edge applications

Zhu

Huang

2018

Trans Emerging Tel Tech

View full text Add to dashboard Cite

show abstract

“…Overloading can result in security attacks, which makes offloading indispensable . Offloading is also essential for achieving lower complexity in cellular networks . Orsini et al studied offloading schemes so that a vast span of services can be sustained.…”

Section: Related Workmentioning

confidence: 99%

E2ARC: Energy‐efficient adaptive resource block allocation with low complexity in device‐to‐device communication

Gandotra

Jha

Jain

2018

Trans Emerging Tel Tech

View full text Add to dashboard Cite

The massive spread of the smart devices demanding high data rate and delay‐sensitized applications has escalated energy consumption levels in the fifth‐generation (5G) wireless communication networks (WCNs). It thus becomes imperative for the mobile network operators to meet the tremendous subscriber demands using energy‐efficient techniques. The limited battery capacities of the mobile terminals (MTs) further make energy efficiency (EE) critical. Device‐to‐device (D2D) communication ensues to be a key enabler technology in the realization of an energy‐efficient 5G WCN. It also serves as a paramount technology for green communication. This paper studies a resource allocation scheme, which iteratively accentuates adaptive reuse of resource blocks (RBs) from the active cellular users to the D2D pairs in a tri‐sectored cell. The cellular users involved in RB allocation to pairs are termed as active. Required count of RBs for each D2D pair is chosen by employing the hidden Markov model. The RB reuse strategy favors enhancement in throughput. The proposed energy‐efficient adaptive RB allocation with low complexity (E2ARC) scheme aims to guarantee ample RBs to individual D2D pairs to optimize the power consumption levels. With advanced throughput and optimal power, augmentation in EE is achieved with a high degree of fairness. E2ARC also comprehends low complexity. The efficacy of the proposed scheme is demonstrated by the simulation results.

show abstract

“…They directly translated input binary to LLVM IRs and thus, do not need to deal with helper function issues. Shuja et al proposed SIMDOM framework . They translated the ARM intrinsic functions to X86 intrinsic functions instead of translating ARM binary to X86 binary.…”

Section: Related Workmentioning

confidence: 99%

Efficient and retargetable SIMD translation in a dynamic binary translator

Hong

Liu

et al. 2018

Softw Pract Exp

View full text Add to dashboard Cite

The single-instruction multiple-data (SIMD) computing capability of modern processors is continually improved to deliver ever better performance and power efficiency. For example, Intel has increased SIMD register lengths from 128 bits in streaming SIMD extension to 512 bits in AVX-512. The ARM scalable vector extension supports SIMD register length up to 2048 bits and includes predicated instructions. However, SIMD instruction translation in dynamic binary translation has not received similar attention. For example, the widely used QEMU emulates guest SIMD instructions with a sequence of scalar instructions, even when the host machines have relevant SIMD instructions. This leaves significant potential for performance enhancement. We propose a newly designed SIMD translation framework for dynamic binary translation, which takes advantage of the host's SIMD capabilities. The proposed framework has been built in HQEMU, an enhanced QEMU with a separate thread for applying LLVM optimizations. The current prototype supports ARMv7, ARMv8, and IA32 guests on the X86-64 AVX-2 host. Compared with the scalar-translation version HQEMU, our framework runs up to 1.84 times faster on Standard Performance Evaluation Corporation 2006 CFP benchmarks and up to 6.81 times faster on selected real applications. FIGURE 3 The architecture of the hybrid QEMU framework. TCG, tiny code generation [Colour figure can be viewed at wileyonlinelibrary.com]

show abstract

SIMDOM: A framework for SIMD instruction translation and offloading in heterogeneous mobile architectures

Cited by 27 publications

References 34 publications

EdgePlace: Availability‐aware placement for chained mobile edge applications

EdgePlace: Availability‐aware placement for chained mobile edge applications

E2ARC: Energy‐efficient adaptive resource block allocation with low complexity in device‐to‐device communication

Efficient and retargetable SIMD translation in a dynamic binary translator

Contact Info

Product

Resources

About