The Importance of Being OS-aware - In Performance Aspects of Cloud Computing Research

  title={The Importance of Being OS-aware - In Performance Aspects of Cloud Computing Research},
  author={Tommaso Cucinotta and Luca Abeni and Mauro Marinoni and Carlo Vitucci},
This paper highlights ineffifiencies in modern cloud infrastructures due to a distance between the research on high-level cloud management / orchestration and the research on low-level kernel and hypervisor mechanisms. Our position about this issue is that more research is needed to make these two worlds talk to each other, providing richer abstractions to describe the low-level mechanisms and automatically map higher-level descriptions and abstractions to configuration and performance tuning… 

Figures from this paper

Virtual Network Functions as Real-Time Containers in Private Clouds
This paper presents preliminary results from our on-going research for ensuring stable performance of co-located distributed cloud services in a resource-efficient way. It is based on using a
Reducing Temporal Interference in Private Clouds through Real-Time Containers
A kernel mechanism that can be applied to the design of an architecture providing fine-grain control of the temporal interferences among concurrent real-time services while avoiding overheads related to machine virtualization is described.
Strong Temporal Isolation among Containers in OpenStack for NFV Services
This paper proposes an approach based on hierarchical real-time CPU scheduling that allows for reserving a precise share of the available computing power for each container deployed in a multi-core server, to provide it with a stable performance, independently from the load of other co-located containers.
Implementation and Deployment of a Server at the Edge Using OpenStack Components
This study uses only a minimal set of Open Stack components in order to evaluate what is the minimal hardware capability needed to set up a basic, but fully working environment for NFV, highlighting the pros and cons of embracing a solution solely based on standard OpenStack components.
The Strategic Role of Inter-Container Communications in RAN Deployment Scenarios
—This paper elaborates on the importance of having ef- ficient inter-container communications at the edge of the network in Software Defined Network – Network Function Virtualization (SDN-NFV)
Allocation and Control of Computing Resources for Real-time Virtual Network Functions
This paper proposes to extend network service descriptors and the Virtualization Infrastructure Manager to enable Network Function Virtualization orchestrators to deploy components exploiting real-time processor scheduling at the hypervisor or host OS level, for enhanced stability of the provided performance.


Real-time cache management for multi-core virtualization
This framework introduces two hypervisor-level techniques, vLLC and vColoring, that enable the cache allocation of individual tasks running in a virtual machine (VM), which is not achievable by the current state of the art.
Improving Responsiveness for Virtualized Networking Under Intensive Computing Workloads
The gathered experimental results show that the proposed technique is effective in controlling the response-times of the real-time workload inside a VM while at the same time it allows for an efficient execution of the batch compute-intensive workload.
Unikernels: library operating systems for the cloud
The Mirage prototype compiles OCaml code into unikernels that run on commodity clouds and offer an order of magnitude reduction in code size without significant performance penalty, and demonstrates that the hypervisor is a platform that overcomes the hardware compatibility issues that have made past library operating systems impractical to deploy in the real-world.
Stochastic Modeling and Performance Analysis of Migration-Enabled and Error-Prone Clouds
This study presents a stochastic-queuing-network-based approach to performance analysis of migration-enabled clouds in error-prone environment and suggests the perfect coverage of theoretical performance results by corresponding experimental confidence intervals.
A cloud-scale acceleration architecture
A new cloud architecture that uses reconfigurable logic to accelerate both network plane functions and applications, and is much more scalable than prior work which used secondary rack-scale networks for inter-FPGA communication.
Comparison of the three CPU schedulers in Xen
This work uses the open source Xen virtual machine monitor to perform a comparative evaluation of three different CPU schedulers for virtual machines and analyzes the impact of the choice of scheduler and its parameters on application performance, and discusses challenges in estimating the application resource requirements in virtualized environments.
Virtualised e-Learning with real-time guarantees on the IRMOS platform
This paper focuses on how Quality of Service guarantees are provided to virtualised applications in the Cloud Computing infrastructure that is being developed in the context of the IRMOS1 European Project and a case-study involving a real e-Learning mobile content delivery application that has been integrated into theIRMOS platform and its achieved performance.
RT-Open Stack: CPU Resource Management for Real-Time Cloud Computing
Experimental results demonstrate that RT-Open Stack can effectively improve the real-time performance of RT VMs while allowing regular VMs to fully utilize the remaining CPU resources.
On the performanc of KVM-based virtual routers
Task-aware virtual machine scheduling for I/O performance.
This paper presents a task-aware virtual machine scheduling mechanism based on inference techniques using gray-box knowledge that infers the I/O-boundness of guest-level tasks and correlates incoming events with I/ O-bound tasks.