Talk:COMP 3000 Essay 2 2010 Question 10 - Revision history

Aellebla: /* Notes to Group */

2010-12-03T06:18:31Z

Notes to Group

← Older revision		Revision as of 06:18, 3 December 2010
Line 2:		Line 2:

	=Notes to Group=		=Notes to Group=
	well i added some more content... and some much needed references, we didn't even have a minimum of three.--[[User:Aellebla\|~~Aellebla~~]] 06:18, 3 December 2010 (UTC)		well i added some more content... and some much needed references, we didn't even have a minimum of three.--[[User:Aellebla\|Aaron Leblanc]] 06:18, 3 December 2010 (UTC)

	Had a bit work on research questions --[[User:xchen6\|xchen6]]		Had a bit work on research questions --[[User:xchen6\|xchen6]]

Aellebla: /* Notes to Group */

2010-12-03T06:18:11Z

Notes to Group

← Older revision		Revision as of 06:18, 3 December 2010
Line 2:		Line 2:

	=Notes to Group=		=Notes to Group=
			well i added some more content... and some much needed references, we didn't even have a minimum of three.--[[User:Aellebla\|Aellebla]] 06:18, 3 December 2010 (UTC)

	Had a bit work on research questions --[[User:xchen6\|xchen6]]		Had a bit work on research questions --[[User:xchen6\|xchen6]]

Aellebla: /* Contribution */

2010-12-03T05:47:35Z

Contribution

← Older revision		Revision as of 05:47, 3 December 2010
Line 92:		Line 92:
	[old]		[old]

	This paper addresses the current limitations of IO resource allocation for hypervisors. The paper has proposed a new and more efficient algorithm to allocate IO resources. Older methods were limited solely by providing proportional shares. mClock incorporates proportional shares, as well as a minimum reservation of IO resources, and a maximum reservation.		This paper addresses the current limitations of IO resource allocation for hypervisors. The paper has proposed a new and more efficient algorithm to allocate IO resources. Older methods were limited solely by providing proportional shares. mClock incorporates proportional shares, as well as a minimum reservation of IO resources, and a maximum reservation.--[[User:Aellebla\|Aaron Leblanc]]

	Older methods of IO resource allocation had a terrible performance lose. Whenever the load on the shared storage device was increased, or when another VM was added, the performance of all hosts would drop considerably. Older methods provided unreliable IO management of hypervisors		Older methods of IO resource allocation had a terrible performance lose. Whenever the load on the shared storage device was increased, or when another VM was added, the performance of all hosts would drop considerably. Older methods provided unreliable IO management of hypervisors--[[User:Aellebla\|Aaron Leblanc]]

	mClock was able to present VMs with a guaranteed minimum reservation of IO resources. This means that application performance will never drop below a certain point. This provides much better application stability on each of the VMs, and better overall performance.		mClock was able to present VMs with a guaranteed minimum reservation of IO resources. This means that application performance will never drop below a certain point. This provides much better application stability on each of the VMs, and better overall performance.--[[User:Aellebla\|Aaron Leblanc]]

	"dmClock (used for cluster-based storage systems) runs a modified version of mClock at each server. There is only one modification to the algorithm to account for the distributed model in the Tag-Assignment component." - from the paper		"dmClock (used for cluster-based storage systems) runs a modified version of mClock at each server. There is only one modification to the algorithm to account for the distributed model in the Tag-Assignment component." - from the paper

Aellebla: /* Research problem */

2010-12-03T05:46:58Z

Research problem

← Older revision		Revision as of 05:46, 3 December 2010
Line 79:		Line 79:


	We use today, a very primitive kind of IO resource allocation in modern hypervisors. Currently an algorithm called PARDA (Proportional Allocation of Resources in Distributed storage Access) 1 is used to allocate IO resources to each VM running on a particular storage device. Unfortunately, the IO resource allocation algorithm of the hosts use a fair-scheduler called SFQ (Start-time Fair Queuing) 2. What this means is that PARDA allocates IO resources to VMs proportional to the number of IO shares on the host, but each host uses a fair scheduler which divides the IO shares amongst the VMs equally. This leads to the problem that whenever another VM is added or another background application is run on one of the VMs, all the other VMs suffer a huge performance lose. In the case of adding another VM, there is a 40% performance drop. This is completely unacceptable when applications have minimum performance requirements to run effectively. An application with minimum resource requirements can be running fine on any given VM, but as soon as the load on the shared storage device increases, the application would run poorly, or could potentially crash. --[[User:Aellebla\|~~Aellebla~~]] ~~05:46, 3 December 2010 (UTC)~~		We use today, a very primitive kind of IO resource allocation in modern hypervisors. Currently an algorithm called PARDA (Proportional Allocation of Resources in Distributed storage Access) 1 is used to allocate IO resources to each VM running on a particular storage device. Unfortunately, the IO resource allocation algorithm of the hosts use a fair-scheduler called SFQ (Start-time Fair Queuing) 2. What this means is that PARDA allocates IO resources to VMs proportional to the number of IO shares on the host, but each host uses a fair scheduler which divides the IO shares amongst the VMs equally. This leads to the problem that whenever another VM is added or another background application is run on one of the VMs, all the other VMs suffer a huge performance lose. In the case of adding another VM, there is a 40% performance drop. This is completely unacceptable when applications have minimum performance requirements to run effectively. An application with minimum resource requirements can be running fine on any given VM, but as soon as the load on the shared storage device increases, the application would run poorly, or could potentially crash. --[[User:Aellebla\|Aaron Leblanc]]

Aellebla: /* Research problem */

2010-12-03T05:46:36Z

Research problem

← Older revision		Revision as of 05:46, 3 December 2010
Line 79:		Line 79:


	We use today, a very primitive kind of IO resource allocation in modern hypervisors. Currently an algorithm called PARDA (Proportional Allocation of Resources in Distributed storage Access) 1 is used to allocate IO resources to each VM running on a particular storage device. Unfortunately, the IO resource allocation algorithm of the hosts use a fair-scheduler called SFQ (Start-time Fair Queuing) 2. What this means is that PARDA allocates IO resources to VMs proportional to the number of IO shares on the host, but each host uses a fair scheduler which divides the IO shares amongst the VMs equally. This leads to the problem that whenever another VM is added or another background application is run on one of the VMs, all the other VMs suffer a huge performance lose. In the case of adding another VM, there is a 40% performance drop. This is completely unacceptable when applications have minimum performance requirements to run effectively. An application with minimum resource requirements can be running fine on any given VM, but as soon as the load on the shared storage device increases, the application would run poorly, or could potentially crash.		We use today, a very primitive kind of IO resource allocation in modern hypervisors. Currently an algorithm called PARDA (Proportional Allocation of Resources in Distributed storage Access) 1 is used to allocate IO resources to each VM running on a particular storage device. Unfortunately, the IO resource allocation algorithm of the hosts use a fair-scheduler called SFQ (Start-time Fair Queuing) 2. What this means is that PARDA allocates IO resources to VMs proportional to the number of IO shares on the host, but each host uses a fair scheduler which divides the IO shares amongst the VMs equally. This leads to the problem that whenever another VM is added or another background application is run on one of the VMs, all the other VMs suffer a huge performance lose. In the case of adding another VM, there is a 40% performance drop. This is completely unacceptable when applications have minimum performance requirements to run effectively. An application with minimum resource requirements can be running fine on any given VM, but as soon as the load on the shared storage device increases, the application would run poorly, or could potentially crash. --[[User:Aellebla\|Aellebla]] 05:46, 3 December 2010 (UTC)

Xchen6: /* Notes to Group */

2010-12-02T22:07:53Z

Notes to Group

← Older revision		Revision as of 22:07, 2 December 2010
Line 2:		Line 2:

	=Notes to Group=		=Notes to Group=
	~~Is it going to be better if we can meet in the lab (or somewhere else) to talk about how we want to work on this essay?~~		Had a bit work on research questions --[[User:xchen6\|xchen6]]
	~~I will be working~~ a bit ~~more~~ on ~~the~~ research ~~problem part, and will be stick to 4175HP for most of the afternoon.~~ --[[User:xchen6\|xchen6]]

	'''We might as well work directly on the main page'''		'''We might as well work directly on the main page'''

Npatel1: /* Notes to Group */

2010-12-02T20:45:35Z

Notes to Group

← Older revision		Revision as of 20:45, 2 December 2010
Line 21:		Line 21:
	: Please edit if you guys have more information or find mistake(s)]--[[User:npatel1\|npatel1]]		: Please edit if you guys have more information or find mistake(s)]--[[User:npatel1\|npatel1]]

	hey, guys! that's what we I've gotten so far on the main page; please check/edit/reference accordingly to essay requirement! ~~(how come two of our other group members did not contribute towards the essay?)~~ --[[User:npatel1\|npatel1]]		hey, guys! that's what we I've gotten so far on the main page; please check/edit/reference accordingly to essay requirement! --[[User:npatel1\|npatel1]]

Xchen6: /* Notes to Group */

2010-12-02T17:40:38Z

Notes to Group

← Older revision		Revision as of 17:40, 2 December 2010
Line 2:		Line 2:

	=Notes to Group=		=Notes to Group=
			Is it going to be better if we can meet in the lab (or somewhere else) to talk about how we want to work on this essay?
			I will be working a bit more on the research problem part, and will be stick to 4175HP for most of the afternoon. --[[User:xchen6\|xchen6]]

	'''We might as well work directly on the main page'''		'''We might as well work directly on the main page'''

Npatel1: /* Notes to Group */

2010-12-02T14:27:10Z

Notes to Group

← Older revision		Revision as of 14:27, 2 December 2010
Line 11:		Line 11:

	I can't any external links on mClock--[[User:tpham\|tpham]]		I can't any external links on mClock--[[User:tpham\|tpham]]


	: Added and made changes on the main page to Background Concepts!		: Added and made changes on the main page to Background Concepts!
Line 17:		Line 18:
	: Added and made changes on the main page to Critique!		: Added and made changes on the main page to Critique!
	: Please edit if you guys have more information or find mistake(s)]--[[User:npatel1\|npatel1]]		: Please edit if you guys have more information or find mistake(s)]--[[User:npatel1\|npatel1]]


	hey, guys! that's what we I've gotten so far on the main page; please check/edit/reference accordingly to essay requirement! (how come two of our other group members did not contribute towards the essay?) --[[User:npatel1\|npatel1]]		hey, guys! that's what we I've gotten so far on the main page; please check/edit/reference accordingly to essay requirement! (how come two of our other group members did not contribute towards the essay?) --[[User:npatel1\|npatel1]]

Npatel1: /* Notes to Group */

2010-12-02T14:26:50Z

Notes to Group

← Older revision		Revision as of 14:26, 2 December 2010
Line 11:		Line 11:

	I can't any external links on mClock--[[User:tpham\|tpham]]		I can't any external links on mClock--[[User:tpham\|tpham]]

			: Added and made changes on the main page to Background Concepts!
			: Added and made changes on the main page to Research Problem!
			: Added and made changes on the main page to Contribution!
			: Added and made changes on the main page to Critique!
			: Please edit if you guys have more information or find mistake(s)]--[[User:npatel1\|npatel1]]


	hey, guys! that's what we I've gotten so far on the main page; please check/edit/reference accordingly to essay requirement! (how come two of our other group members did not contribute towards the essay?) --[[User:npatel1\|npatel1]]		hey, guys! that's what we I've gotten so far on the main page; please check/edit/reference accordingly to essay requirement! (how come two of our other group members did not contribute towards the essay?) --[[User:npatel1\|npatel1]]


	Hey, so I think tag assignment, tag adjustment, and request scheduler might be something important but i don't quite understand it if someone can take a look at it and see what they can come up with that would be helpful its on page 5 --[[User:tpham\|tpham]]		Hey, so I think tag assignment, tag adjustment, and request scheduler might be something important but i don't quite understand it if someone can take a look at it and see what they can come up with that would be helpful its on page 5 --[[User:tpham\|tpham]]