Cluster-Wide Context Switch of Virtualized Jobs Fabien Hermenier, Adrien Lèbre, Jean-Marc Menaud
VTDC’10, 22 June 2010
ASCOLA Team
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
1 / 23
Agenda Motivation Global Design Architecture Implementation Proof of concept A sample scheduler Experiment on a cluster Conclusion
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
2 / 23
Motivation
Agenda Motivation Global Design Architecture Implementation Proof of concept A sample scheduler Experiment on a cluster Conclusion
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
3 / 23
Motivation
Motivation
Clusters I
large infrastructures to execute various jobs
Resource Management System (RMS) I
manage the execution of jobs
I
resources are allocated to jobs according to their description
I
scheduling: which jobs to execute, and where ?
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
4 / 23
Motivation
Jobs schedulers Usually A corse-grain exploitation of resources : I
static allocation of resources
I
execution to completion
Dynamic schedulers exist Based on mechanisms that manipulate the jobs dynamically (migration, preemption, dynamic allocation of resources, . . . ). BUT I
mechanisms are complex to implement
I
mechanisms are complex to use efficiently
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
5 / 23
Motivation
Motivation Virtual Machines (VMs) as a backend for dynamic schedulers I
each component is embedded into its VM
I
VMMs provide migration, preemption
I
still complex to use efficiently
A cutting-edge building block dynamic consolidation, best-effort jobs , . . . I I
various policies, but common concepts to perform the changes each provides an ad-hoc solution to handle several common issues: I I I
dependencies between actions correctness reactivity
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
6 / 23
Motivation
Proposition Performing the changes should not be a primary concern for developers I
a generic cluster-wide context switch based on VMs
I
developers only focus on the algorithm to select the jobs to run the cluster-wide context switch takes care of the rest
I
I I I
detects the changes to perform ensures the correctness of the transition computes the fastest possible transition
The implementation leverages the consolidation manager Entropy
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
7 / 23
Global Design
Agenda Motivation Global Design Architecture Implementation Proof of concept A sample scheduler Experiment on a cluster Conclusion
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
8 / 23
Global Design
Architecture
From jobs to virtualized Jobs
Figure: The life cycle of a vjob
I
a vjob encapsulates one or several VMs
I
to change the state of a vjob, actions (except migrate) are executed on each VMs
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
9 / 23
Global Design
Architecture
Configuration I
describes the assignment of the running VMs to working nodes
I
nodes provide CPU and memory resources
I
running VMs require CPU and memory resources to run at peak level
(a) Non-viable configuration
Hermenier et al.
(ASCOLA)
(b) Viable configuration
Cluster-Wide Context Switch of Virtualized Jobs
10 / 23
Global Design
Architecture
The control loop of Entropy
Monitor I
extract the current configuration: VM position, CPU/memory consumption
I
adaptable to a specific monitoring system (currently Ganglia)
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
11 / 23
Global Design
Architecture
The control loop of Entropy
Scheduling policy I
an algorithm to select the vjobs to run wrt. the current configuration
I
provided by a developer
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
11 / 23
Global Design
Architecture
The control loop of Entropy
The cluster-wide context switch module I
selects a position for each VM to run
I
infers the actions that make the transition w. the current configuration
I
computes the fastest plan that ensure the correctness of the process
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
11 / 23
Global Design
Architecture
The control loop of Entropy
Execution I
associate each action of the plan with a driver that performs the action
I
adaptable to specific environments. Currently support Xen VMM (XML-RPC) or shell command
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
11 / 23
Global Design
Implementation
Role of the CW context switch
I
detects the actions to perform
I
selects a position for each VM to run
I
plans the actions to guarantee the correctness of the process
I
computes the fastest possible plan
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
12 / 23
Global Design
Implementation
Plan the actions
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
13 / 23
Global Design
Implementation
Plan the actions
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
13 / 23
Global Design
Implementation
Plan the actions
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
13 / 23
Global Design
Implementation
Plan the actions
The reconfiguration plan I
a protocol to execute actions
I
actions feasible in parallel are grouped into a same step
I
steps are executed sequentially
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
13 / 23
Global Design
Implementation
Suspending/Resuming a vjob I
inter-connected VMs should be continuously in the same state
I
coordination to ensure that distributed applications will not fail
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
14 / 23
Global Design
Implementation
Suspending/Resuming a vjob I I
I I
inter-connected VMs should be continuously in the same state coordination to ensure that distributed applications will not fail
actions are grouped into a same step synchronization between the pause/unpause actions
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
14 / 23
Global Design
Implementation
Reducing the duration of a cluster-wide context switch
45
start/run migrate stop/shutdown
40
local nfs localïscp localïrsync
200
150
30
Completion time (in sec)
Completion time (in sec)
35
25 20 15 10
100
50
5 0 128 256
512
1024 VM size (in MB)
2048
0 128 256
512
1024 VM size (in MB)
I
the duration of an action depends on its context
I
a function estimates the cost of a whole CW context switch
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
2048
15 / 23
Global Design
Implementation
Reducing the duration of a CW context switch An approach based on constraint programing Entropy computes a new configuration that I
is viable
I
respects the scheduling policy
I
implies the minimal cost
In practice I
actions are performed asap.
I
prefer moving VMs will small memory requirements
I
avoid migrations and remote resumes
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
16 / 23
Proof of concept
Agenda Motivation Global Design Architecture Implementation Proof of concept A sample scheduler Experiment on a cluster Conclusion
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
17 / 23
Proof of concept
A sample scheduler
A sample scheduler Principle I
a FIFO queue
I
VMs are assigned to nodes using a First Fit Decrease heuristic
I
priority between jobs to prevent starvation
Example
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
18 / 23
Proof of concept
A sample scheduler
A sample scheduler Principle I
a FIFO queue
I
VMs are assigned to nodes using a First Fit Decrease heuristic
I
priority between jobs to prevent starvation
Example
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
18 / 23
Proof of concept
A sample scheduler
A sample scheduler Principle I
a FIFO queue
I
VMs are assigned to nodes using a First Fit Decrease heuristic
I
priority between jobs to prevent starvation
Benefits using CW context switch I
dynamic allocation of resources
I
preemption
I
migration of VMs
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
18 / 23
Proof of concept
Experiment on a cluster
Environment Hardware I
11 working nodes
I
3 storage nodes share VM images
I
1 service node is running Entropy
Protocol I
a queue of 8 vjobs (NASGrid benchmarks)
I
each vjob uses 9 VMs comparison with regards to FCFS
I
I I
resources usage completion time
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
19 / 23
Proof of concept
Experiment on a cluster
Experiment on a cluster Benefits I
improve resource usage
I
suspend/resume transparent for the developer
Resources usage
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
20 / 23
Proof of concept
Experiment on a cluster
Experiment on a cluster
Benefits I
improve resource usage
I
suspend/resume transparent for the developer
I
reduce the completion time
Cumulated execution time I
FCFS: 250 minutes
I
Entropy: 150 minutes
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
20 / 23
Conclusion
Agenda Motivation Global Design Architecture Implementation Proof of concept A sample scheduler Experiment on a cluster Conclusion
Hermenier et al.
(ASCOLA)
Cluster-Wide Context Switch of Virtualized Jobs
21 / 23
Conclusion
Conclusion RMSs start to manage VMs instead of process I
VMMs provide mechanisms to implement dynamic schedulers
I
manipulate VMs is tedious and may be non cost-effective
I
various scheduling policies but common concepts to perform the context switch
A generic cluster-wide context switch I
make the implementation of dynamic schedulers easier
I
the context switch is outside the scheduling algorithm
I
an implementation in Entropy with a sample algorithm http://entropy.gforge.inria.fr version 1.2 (LGPL)
resources are allocated to jobs according to their description. â· scheduling: ..... fond of - virtualization, distributed systems, autonomic computing, . . . â· dislike - ...
developers only focus on the algorithm to select the jobs to run. â· the cluster-wide context switch takes care of the rest. â» detects the changes to perform.
Architecture. Implementation ... The implementation leverages the consolidation manager Entropy .... actions feasible in parallel are grouped into a same step.
new tools for predicting performance, providing information for resource alloca- .... In other words, models depend on data collected by measurement tools in ...
vendors has been to sell the necessary software, equip- ment, and appliance nodes to realize and manage large-scale ... high volume servers, storage devices, and switches. In this paper we focus on Network Function .... Abstract Network Application o
WinPak is a software that controls access control to all of our building door lock systems. This service used to run on ... Pittsburg Community Schools Mail - Virtualized Server Host Lists ... When the 4th physical (ESX) Host is purchased for the SAN
circuit and relay will be activated. As Capacitance Level Switch has no moving parts inside the device, it will not be affected by friction. It is suitable for powder or liquid application easy to install. The customer can choose the types for his re
timer in clockwise. The relay will energized after. "Indicator" illuminate for several seconds if set timer more than 0 second. The delay function is suitable for ...
These are mostly SMD components. The processor is in the THT, which will. facilitate the exchange of. Page 3 of 10. SWITCH eng.pdf. SWITCH eng.pdf. Open.
#define second_led 13 â//second LED is connected to 13th pin. #define first_datapin 10 â//D1 of slide switch is connected to. 10th pin. #define second_datapin ...
Dec 25, 2012 - ee app lcanon e or Comp ete Seam lstory' is actuated by light of su?icient .... Will be used to make this calculation. Where no is 4 pi>
First, CPU accounting in guest OSes can be inaccurate under discontinuous time, leading to false identi- cation of I/O-bound task as compute-bound. Second and most importantly, work-conserving (WC) scheduling, which is designed for continuous CPU ava
machinery that draws both on the ideas of the anti-formalist GriceâAustin tradition .... the time of utterance, this will not be true for answering machine messages.
Apr 1, 2013 - assistance income) and unstable neighborhoods (higher proportions of 1-person ... With this knowledge, city planners and homeless service providers can better use limited ..... 1996;7(2):327-365. 15. ... [computer program].
adaptation on a large vocabulary mobile speech recognition task. Index Termsâ Large ... estimated directly from the speaker data, but using the well-trained speaker ... quency ceptral coefficients (MFCC) or perceptual linear prediction. (PLP) featu
the information on which context-dependent speech acts depend, and the situation that speech acts ..... The other was in fact the Secretary of Health and Human.
PDF Microsoft System Center: Building a. Virtualized Network Solution Full Books. Books detail. Title : PDF Microsoft System Center: Building a q. Virtualized ...
Aug 5, 2016 - tracing mechanisms and this calls for a system level and application transparent tracing tool. There exist ... trace network latency at packet level in virtualized environ- ments. TC timestamps packets at ..... its fair CPU share, it al