A case for RDMA in clouds

Trivedi, Animesh; Metzler, Bernard; Stuedi, Patrick

doi:10.1145/2103799.2103820

Cited by 23 publications

(3 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These include InfiniBand, as well as 10Gb and 40Gb Ethernet [14,15]. To make use of their high throughput, they rely on RDMA [16]. Variants are summarized by Huang et al [17] and include native RDMA over InfiniBand, Converged Enhanced Ethernet (RoCE), and Internet Wide Area RDMA Protocol (iWARP).…”

Section: Distributed I/o Using Rdmamentioning

confidence: 99%

Flexible device compositions and dynamic resource sharing in PCIe interconnected clusters using Device Lending

et al. 2019

View full text Add to dashboard Cite

Modern workloads often exceed the processing and I/O capabilities provided by resource virtualization, requiring direct access to the physical hardware in order to reduce latency and computing overhead. For computers interconnected in a cluser, access to remote hardware resources often requires facilitation both in hardware and specialized drivers with virtualization support. This limits the availability of resources to specific devices and drivers that are supported by the virtualization technology being used, as well as what the interconnection technology supports. For PCI Express (PCIe) clusters, we have previously proposed Device Lending as a solution for enabling direct low latency access to remote devices. The method has extremely low computing overhead, and does not require any application- or device-specific distribution mechanisms. Any PCIe device, such as network cards disks, and GPUs, can easily be shared among the connected hosts. In this work, we have extended our solution with support for a virtual machine (VM) hypervisor. Physical remote devices can be “passed through” to VM guests, enabling direct access to physical resources while still retaining the flexibility of virtualization. Additionally, we have also implemented multi-device support, enabling shortest-path peer-to-peer transfers between remote devices residing in different hosts.Our experimental results prove that multiple remote devices can be used, achieving bandwidth and latency close to native PCIe, and without requiring any additional support in device drivers. I/O intensive workloads run seamlessly using both local and remote resources. With our added VM and multi-device support, Device Lending offers highly customizable configurations of remote devices that can be dynamically reassigned and shared to optimize resource utilization, thus enabling a flexible composable I/O infrastructure for VMs as well as bare-metal machines.

show abstract

Section: Distributed I/o Using Rdmamentioning

confidence: 99%

Flexible device compositions and dynamic resource sharing in PCIe interconnected clusters using Device Lending

et al. 2019

View full text Add to dashboard Cite

show abstract

“…These include InfiniBand, as well as 10Gb and 40Gb Ethernet [5,16]. To make use of their high throughput, they rely on RDMA [29]. Variants are summarized by Huang et al [12] and include native RDMA over InfiniBand, Converged Enhanced Ethernet (RoCE), and Internet Wide Area RDMA Protocol (iWARP).…”

Section: Distributed Io Using Rdmamentioning

confidence: 99%

Flexible Device Sharing in PCIe Clusters using Device Lending

Markussen

Kristiansen

Stensland

et al. 2018

Proceedings of the 47th International Conference on Parallel Processing Companion

View full text Add to dashboard Cite

Processing workloads may have very high IO demands, exceeding the capabilities provided by resource virtualization and requiring direct access to the physical hardware. For computers that are interconnected in PCI Express (PCIe) networks, we have previously proposed Device Lending as a solution for assigning devices to remote hosts. In this paper, we explain how we have extended our implementation with support for the Linux Kernel-based Virtual Machine (KVM) hypervisor. Using our extended Device Lending, it becomes possible to dynamically "pass through" physical remote devices to VM guests while still retaining the flexibility of virtualization, something that previously required extensive facilitation in both hypervisor and device drivers in the form of paravirtualization. We have also improved our original implementation with support for interoperability between remote devices. We show that it is possible to use multiple devices residing in different hosts, while still achieving the same bandwidth and latency as native PCIe, and without requiring any additional support in device drivers. CCS CONCEPTS • Computer systems organization → Distributed architectures; Interconnection architectures; Cloud computing;

show abstract

“…RDMA is a standard communication model that allows a process to directly access the memory address space of another process on a remote machine [33]. The RDMA specification defines an asynchronous, general-purpose semantics to leverage the features of network accelerators, such as zero-copy communication, but it does not require a specific implementation: any hardware device, including in principle general-purpose CPUs [82], may implement its model. Thus, RDMA introduces a uniform access layer to hardware acceleration.…”

Section: B Introduction To Rdmamentioning

confidence: 99%

Empowering Cloud Computing With Network Acceleration: A Survey

Rosa,

Foschini,

Corradi

2024

IEEE Commun. Surv. Tutorials

View full text Add to dashboard Cite

Modern interactive and data-intensive applications must operate under demanding time constraints, prompting a shift toward the adoption of specialized software and hardware network acceleration technologies. This specialization, however, poses significant scalability, flexibility, security, and economic sustainability challenges for application developers. Cloud computing holds the potential to overcome these obstacles by offering the cost-effective option to access specialized acceleration technologies through standard cloud interfaces. Nevertheless, that integration is still challenging for cloud providers. In the cloud, physical resources are hidden behind a virtualization layer, whereas acceleration technologies make applications directly interact with the hardware. To bridge this gap, recent literature explores the possibility of empowering cloud platforms with accelerated networking as a commodity, thus offering the innovative option of Network Acceleration as a Service. This survey reviews these recent research efforts by adopting popular technologies like XDP, DPDK, and RDMA as a reference. To organize the surveyed research in a comprehensive framework, we identify four key aspects that pose critical problems to the integration of acceleration options in cloud computing -access interfaces, virtualization techniques, serviceability, and security -and systematically discuss the associated challenges. Then, we present the issues to be further addressed and outline the most promising research directions for the full integration of network acceleration within next-generation cloud computing platforms.

show abstract

A case for RDMA in clouds

Cited by 23 publications

References 17 publications

Flexible device compositions and dynamic resource sharing in PCIe interconnected clusters using Device Lending

Flexible device compositions and dynamic resource sharing in PCIe interconnected clusters using Device Lending

Flexible Device Sharing in PCIe Clusters using Device Lending

Empowering Cloud Computing With Network Acceleration: A Survey

Contact Info

Product

Resources

About