Archived/GSoC 2012 Ideas: Difference between revisions

From Xen
Jump to navigationJump to search
Line 66: Line 66:
* Kang Hua <kanghua151@gmail.com>
* Kang Hua <kanghua151@gmail.com>
| Size: Medium. Knowledge: C, block driver, File system and HDFS
| Size: Medium. Knowledge: C, block driver, File system and HDFS
| HLFS, sub-project of Cloudxy, is a distributed vm image storage system for XEN. HLFS(Hadoop Log Structure Filesystem,actually named Hadoop Log Structure Block System is more proper) provides highly available block level storage volumes that can be attached to XEN virtual machines by its tapdisk driver — Similar projects in KVM region is sheepdog,however in different architecture. For details, you can visit http://code.google.com/p/cloudxy/wiki/WHAT_IS_CLOUDXY or http://code.google.com/p/cloudxy (In Cheinese).
| HLFS, sub-project of Cloudxy, is a distributed vm image storage system for XEN. HLFS(Hadoop Log Structure Filesystem,actually named Hadoop Log Structure Block System is more proper) provides highly available block level storage volumes that can be attached to XEN virtual machines by its tapdisk driver — Similar projects in KVM region is sheepdog,however in different architecture. For details, you can visit http://code.google.com/p/cloudxy/wiki/WHAT_IS_CLOUDXY or http://code.google.com/p/cloudxy (In Chinese).
We hope to accomplish two features, which are "clean segment under tree snapshots" feature and "block cache and write-back" feather during GSOC 2012.<br>
We hope to accomplish two features, which are "clean segment under tree snapshots" feature and "block cache and write-back" feather during GSOC 2012.<br>
The first one will involve:
The first one will involve:
Line 79: Line 79:
* Harry Wei <harryxiyou@gmail.com>
* Harry Wei <harryxiyou@gmail.com>
* Kelvin <senwang@linux.vnet.ibm.com>
* Kelvin <senwang@linux.vnet.ibm.com>
* Zhang Zhang <zhang_zhang@live.com>
|}
|}



Revision as of 08:00, 11 March 2012

Project Ideas for GSoC 2012

This page is used to list project ideas for Google Summer of Code (GSOC) 2012. Xen will apply as a mentoring organization again.

Key Google Pages

banner-gsoc2012.png

Project Ideas

Rules

  • Be creative
  • Be specific: what do you want to be implemented; if at all possible provide an indication of size and complexity as described above to make it easier for a student to choose ideas
  • If you are willing to mentors those ideas, add your name and email to the idea.
  • If you're an interested student, add your name and email next to the idea. It is ok to have several students interested by one idea.
  • Aspiring students need to get in touch with the xen.org community manager via community.manager@xen.org to register their interest

New Project Ideas

Please add new ideas with a description or mail community.manager@xen.org if in doubt. - This is now open for 2012 GSoC!

Description Project Possible Mentor(s) Size, Knowledge required Detailed Description Possible Student(s)
<Your Idea> <Your Project; e.g. Xen or XCP or XenARM or... > <mentor contact info> <skills required: e.g. Qemu, Xen, Linux> <description of the project idea> <student contact info>
TRIM support for xen virtual block device blktap2, xenblk George Shuklin <shuklin@selectel.ru> Size:small, Linux block drivers, generic system programming If xenblk supported TRIM operation, it would've allowed to let backend to know about 'unused space'. In this case thin provision whould be bi-directional, providing way to 'shrink' overgrown thin-provisioned disk of virtual machine.
Port XCP Toolstack to ARM (unconfirmed) XCP Mike McClurg <mike.mcclurg@citrix.com> Size: Medium/Large. Knowledge: Debian packaging; compiling for Linux on ARM; basic C, Python and OCaml A team of core Xen engineers has just finished porting the Xen hypervisor to the unreleased ARM A15 platform. In preparation for new A15 hardware, we'd like to make sure that the XCP toolstack runs on on current ARM hardware in emulated-Xen mode. This project would involve modifying the XCP toolstack build system to run on current ARM hardware, and produce an XCP toolstack that will run in emulated-Xen mode.
Virtulazation of the CPU Performance Monitoring Unit (unconfirmed) XCP Marcus Granado <marcus.granado@citrix.com> Size: Unknown. Skills: C, kernel/hypervisor hackery Right now, the CPU PMU (performance monitoring unit) is not virtualized, therefore in general it is not possible to run hardware counter profilers (such as Linux oprofile, FreeBSD hwpmc or similar tools for Windows) inside guest domains. The goal of the project would be to allow these hardware counter profilers to run inside guest domains.
Improvement of OpenXenManager, the XCP GUI XCP Mike McClurg <mike.mcclurg@citrix.com> Size: Medium. Knowledge: Python and GTK OpenXenManager is a free, open-source, multi-platform GUI for both Xen Cloud Platform and XenServer. It has many nice features, but could use some bug fixes and new features.
"Segment Clean" & "Cache" feather of HLFS which is sub-project of Cloudxy (unconfirmed) XEN,BLKTAP2,Hadoop
  • Chen Lijun <cljcore@126.com>
  • Kang Hua <kanghua151@gmail.com>
Size: Medium. Knowledge: C, block driver, File system and HDFS HLFS, sub-project of Cloudxy, is a distributed vm image storage system for XEN. HLFS(Hadoop Log Structure Filesystem,actually named Hadoop Log Structure Block System is more proper) provides highly available block level storage volumes that can be attached to XEN virtual machines by its tapdisk driver — Similar projects in KVM region is sheepdog,however in different architecture. For details, you can visit http://code.google.com/p/cloudxy/wiki/WHAT_IS_CLOUDXY or http://code.google.com/p/cloudxy (In Chinese).

We hope to accomplish two features, which are "clean segment under tree snapshots" feature and "block cache and write-back" feather during GSOC 2012.
The first one will involve: 1. Implement mechanism of segment clean - it involves "mid-snapshot segment clean" and "over-snapshot segment clean". so you can do it in normal way(pull data) and "map/reduce" way. 2. Implement strategy of segment clean - it will take into account that how to avoid IO contention between clean operation and vm disk access operation. You can visit http://code.google.com/p/cloudxy/wiki/HlfsDesign_English_Version for more details about HLFS's realize mechanisms.
Another one will involve: 1. Read and write cache management - include cache replacement mechanism. 2. Write-back management - include gather more block to write for reducing write operations.

  • Harry Wei <harryxiyou@gmail.com>
  • Kelvin <senwang@linux.vnet.ibm.com>
  • Zhang Zhang <zhang_zhang@live.com>