Re: [squid-users] Re: Re: [squid-users] centralized storage for squid
2008/3/11 Neil Harkins [EMAIL PROTECTED]: F5 has some documents on how to implement consisent hashes in bigip irules (tcl), but i wound up writing a custom one for use in front of our squids that only does one checksum per request, as opposed to one per squid in the pool, to avoid wasting cpu cycles on the LB. it uses a precomputed table for the nodes, but doesn't need to be recomputed when you add/remove a few, they just fit in between the others. i'll try to finish the writeup and submit it to devcentral soon. You're also welcome to write about this issue in the squid wiki (be it a link to devcentral or an article in the wiki itself) -- /kinkie
Re: [squid-users] Re: Re: [squid-users] centralized storage for squid
F5 has some documents on how to implement consisent hashes in bigip irules (tcl), but i wound up writing a custom one for use in front of our squids that only does one checksum per request, as opposed to one per squid in the pool, to avoid wasting cpu cycles on the LB. it uses a precomputed table for the nodes, but doesn't need to be recomputed when you add/remove a few, they just fit in between the others. i'll try to finish the writeup and submit it to devcentral soon. -neil 2008/3/10 Mark Nottingham [EMAIL PROTECTED]: This is the problem that CARP and other consistent hashing approaches are supposed to solve. Unfortunately, the Squid in the front will often be a bottleneck... Cheers, On 07/03/2008, at 1:43 PM, Siu Kin LAM wrote: Hi Pablo Actually, it is my case. The URL-hash is helpful to reduce the duplicated objects. However, once adding/removing squid server, load balancer needs to re-calculate the hash of URL which cause lot of TCP_MISS in squid server at the inital stage. Do you have same experience ? Thanks --- Pablo Garc燰 [EMAIL PROTECTED] 說: I dealt with the same problem using a load balancer in front of the cache farm, using a URL-HASH algorithm to send the same url to the same cache every time. It works great, and also increases the hit ratio a lot. Regards, Pablo 2008/3/6 Siu Kin LAM [EMAIL PROTECTED]: Dear all At this moment, I have several squid servers for http caching. Many duplicated objects have been found in different servers. I would minimize to data storage by installing a large centralized storage and the squid servers mount to the storage as data disk. Have anyone tried this before? thanks a lot Yahoo! 網上安全攻略,教你如何防範黑客! 請前往http://hk.promo.yahoo.com/security/index.html 了解更多。 Yahoo! 網上安全攻略,教你如何防範黑客! 請前往http://hk .promo.yahoo.com/security/index.html 了解更多。 -- Mark Nottingham [EMAIL PROTECTED]
Re: [squid-users] Re: Re: [squid-users] centralized storage for squid
I have the same problem, tough I don't remove my squid servers very often. I've partially resolved this problem thanks to the implementation of the algorithm of my load balancer. What it does, it's calculate the hash taking into account all the squids in the pool, whether they're up or down. Then if the algorithm chooses a server that is down, then the calculation happens again. So if one of my squids restarts, as soon as it starts again, it receives the same urls as before using disk cache instead of memory. What you also can try is to link all the squids toghether whith icp to create a sibling relationship between them. though I guess the hierarchical cache scenario would help you best to reduce the load in your web servers. Hope this helps, Regards, Pablo 2008/3/7 Siu Kin LAM [EMAIL PROTECTED]: Hi Pablo Actually, it is my case. The URL-hash is helpful to reduce the duplicated objects. However, once adding/removing squid server, load balancer needs to re-calculate the hash of URL which cause lot of TCP_MISS in squid server at the inital stage. Do you have same experience ? Thanks --- Pablo Garc燰 [EMAIL PROTECTED] 說: I dealt with the same problem using a load balancer in front of the cache farm, using a URL-HASH algorithm to send the same url to the same cache every time. It works great, and also increases the hit ratio a lot. Regards, Pablo 2008/3/6 Siu Kin LAM [EMAIL PROTECTED]: Dear all At this moment, I have several squid servers for http caching. Many duplicated objects have been found in different servers. I would minimize to data storage by installing a large centralized storage and the squid servers mount to the storage as data disk. Have anyone tried this before? thanks a lot Yahoo! 網上安全攻略,教你如何防範黑客! 請前往http://hk.promo.yahoo.com/security/index.html 了解更多。 Yahoo! 網上安全攻略,教你如何防範黑客! 請前往http://hk.promo.yahoo.com/security/index.html 了解更多。