Arrakis: The Operating System is the Control Plane

@inproceedings{Peter2013ArrakisTO,
  title={Arrakis: The Operating System is the Control Plane},
  author={Simon Peter and Jialin Li and Irene Zhang and Dan R. K. Ports and Doug Woos and Arvind Krishnamurthy and Thomas E. Anderson and Timothy Roscoe},
  booktitle={OSDI},
  year={2013}
}
Recent device hardware trends enable a new approach to the design of network server operating systems. In a traditional operating system, the kernel mediates access to device hardware by server applications, to enforce process isolation as well as network and disk security. We have designed and implemented a new operating system, Arrakis, that splits the traditional role of the kernel in two. Applications have direct access to virtualized I/O devices, allowing most I/O operations to skip the… 

Figures and Tables from this paper

We need kernel interposition over the network dataplane
TLDR
This work hypothesizes that KOPI can support the same tools and policies as the kernel stack while retaining the performance benefits of kernel bypass, and proposes Kernel On-Path Interposition (KOPI), in which traditional kernel data-plane functionality is retained but implemented in a fully programmable SmartNIC.
I'm Not Dead Yet!: The Role of the Operating System in a Kernel-Bypass Era
TLDR
The Demikernel is proposed, a new library OS architecture for kernel-bypass devices that defines a high-level, kernel-Bypass I/O abstraction and provides user-space library OSes to implement that abstraction across a range of kernel- bypass devices.
The eXpress data path: fast programmable packet processing in the operating system kernel
TLDR
It is shown that XDP achieves single-core packet processing performance as high as 24 million packets per second, and illustrates the flexibility of the programming model through three example use cases: layer-3 routing, inline DDoS protection and layer-4 load balancing.
XOS: An Application-Defined Operating System for Data Center Servers
TLDR
XOS is presented, an application-defined OS for modern DC servers that leverages modern hardware support for virtualization to move resource management functionality out of the conventional kernel and into user space, which lets applications achieve near bare-metal performance.
IX: A Protected Dataplane Operating System for High Throughput and Low Latency
TLDR
IX is presented, a dataplane operating system that provides high I/O performance, while maintaining the key advantage of strong protection offered by existing kernels, and outperforms Linux and state-of-the-art, user-space network stacks significantly in both throughput and end-to-end latency.
NeSC: Self-virtualizing nested storage controller
TLDR
The self-virtualizing, nested storage controller (NeSC) architecture is presented, which includes a filesystem-agnostic protection mechanism that enables the physical device to export files as virtual PCIe storage devices and offloads the hypervisor's storage layer functionality to hardware.
CrossFS: A Cross-layered Direct-Access File System
TLDR
Current User- FS, Kernel-FS, and Firmware-FS designs lack a synergistic design across the user, the kernel, and the firmware layers, which is critical for achieving direct storage access and scaling concurrent I/O performance without compromising fundamental file system properties.
Hodor: Intra-Process Isolation for High-Throughput Data Plane Libraries
TLDR
This paper proposes protected libraries as a new OS abstraction which provides separate user-level protection domains for different services, with performance approaching that of unprotected kernel bypass, and shows that this approach can efficiently protect highthroughput in-memory databases and user-space network stacks.
User-Space I/O for µ s-level Storage Devices
TLDR
Iris, a new I/O path for applications, that minimizes overheads from system software in the common I/ O path, is presented and it achieves on a single core up to \(1.7\times\) and \(2.2\times \) better read and write random IOPS, respectively, compared to the xfs and ext4 file systems.
XOS: An Application-Defined Operating System for Datacenter Computing
TLDR
XOS is presented, an application-defined OS for modern DC servers that leverages modern hardware support for virtualization to move resource management functionality out of the conventional kernel and into user space, which lets applications achieve near bare-metal performance.
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 113 REFERENCES
Exokernel: an operating system architecture for application-level resource management
TLDR
The prototype exokernel system implemented here is at least five times faster on operations such as exception dispatching and interprocess communication, and allows applications to control machine resources in ways not possible in traditional operating systems.
IX: A Protected Dataplane Operating System for High Throughput and Low Latency
TLDR
IX is presented, a dataplane operating system that provides high I/O performance, while maintaining the key advantage of strong protection offered by existing kernels, and outperforms Linux and state-of-the-art, user-space network stacks significantly in both throughput and end-to-end latency.
Extensibility safety and performance in the SPIN operating system
This paper describes the motivation, architecture and performance of SPIN, an extensible operating system. SPIN provides an extension infrastructure, together with a core set of extensible services,
Arrakis: A Case for the End of the Empire
In this paper, we argue that recent device hardware trends enable a new approach to the design of operating systems: instead of the operating system mediating access to hardware, applications run
Dune: Safe User-level Access to Privileged CPU Features
TLDR
This work uses Dune to implement three user-level applications that can benefit from access to privileged hardware: a sandbox for untrusted code, a privilege separation facility, and a garbage collector, and greatly simplifies the implementation of these applications and provides significant performance advantages.
IO-Lite: a unified I/O buffering and caching system
TLDR
IO-Lite eliminates all copying and multiple buffering of I/O data, and enables various cross-subsystem optimizations, and shows performance improvements between 40 and 80% on real workloads as a result of IO -Lite.
FlexNIC: Rethinking Network DMA
TLDR
FlexNIC, a flexible network DMA interface that can be used by operating systems and applications alike to reduce packet processing overheads, is proposed and shown how it can benefit widely used data center server applications, such as key-value stores.
netmap: A Novel Framework for Fast Packet I/O
  • L. Rizzo
  • Computer Science
    USENIX Annual Technical Conference
  • 2012
TLDR
The novelty in the proposal is not only that it exceeds the performance of most of previous work, but also that it provides an architecture that is tightly integrated with existing operating system primitives, not tied to specific hardware, and easy to use and maintain.
Providing safe, user space access to fast, solid state disks
TLDR
This work describes a novel storage hardware and software architecture that nearly eliminates two sources of overhead: Entering the kernel and performing file system permission checks and provides a private, virtualized interface for each process and moves file system protection checks into hardware.
Protocol implementation in a vertically structured operating system
TLDR
This paper describes the techniques used in the protocol implementation and the benefits that the vertical structure provided and the techniques they were used to achieve.
...
1
2
3
4
5
...