search for: 175ms

...t the overhead is. > Hi MST, I have measured the performance when using a 32K page bitmap, and inflate the balloon to 3GB of an idle guest with 4GB RAM. Now: total inflating time: 338ms the count of virtio data transmission: 373 the call count of madvise: 865 before: total inflating time: 175ms the count of virtio data transmission: 1 the call count of madvise: 42 Maybe the result will be worse if the guest is not idle, or the guest has more RAM. Do you want more data? Is it worth to do that? Liang > > > > > 2. limit allocated bitmap size to something reasonable. >...

...t the overhead is. > Hi MST, I have measured the performance when using a 32K page bitmap, and inflate the balloon to 3GB of an idle guest with 4GB RAM. Now: total inflating time: 338ms the count of virtio data transmission: 373 the call count of madvise: 865 before: total inflating time: 175ms the count of virtio data transmission: 1 the call count of madvise: 42 Maybe the result will be worse if the guest is not idle, or the guest has more RAM. Do you want more data? Is it worth to do that? Liang > > > > > 2. limit allocated bitmap size to something reasonable. >...

...--------------------------------------------------------------------------------- Because the PFNs of the allocated pages are not linear increased, so 3 transmissions are impossible. Liang > > > the call count of madvise: 865 > > > > before: > > total inflating time: 175ms > > the count of virtio data transmission: 1 the call count of madvise: 42 > > > > Maybe the result will be worse if the guest is not idle, or the guest has > more RAM. > > Do you want more data? > > > > Is it worth to do that? > > > > Liang >...

...--------------------------------------------------------------------------------- Because the PFNs of the allocated pages are not linear increased, so 3 transmissions are impossible. Liang > > > the call count of madvise: 865 > > > > before: > > total inflating time: 175ms > > the count of virtio data transmission: 1 the call count of madvise: 42 > > > > Maybe the result will be worse if the guest is not idle, or the guest has > more RAM. > > Do you want more data? > > > > Is it worth to do that? > > > > Liang >...

...the overhead of stage c and stage d can also be reduced a lot. This patch is the kernel side implementation which is intended to speed up the inflating & deflating process by adding a new feature to the virtio-balloon device. And now, inflating the balloon to 3GB of a 4GB idle guest only takes 175ms, it's about 9 times as fast as before. TODO: optimize stage a by allocating/freeing a chunk of pages instead of a single page at a time. Signed-off-by: Liang Li <liang.z.li at intel.com> --- drivers/virtio/virtio_balloon.c | 199 ++++++++++++++++++++++++++++++++++-- include/uapi/li...

...the overhead of stage c and stage d can also be reduced a lot. This patch is the kernel side implementation which is intended to speed up the inflating & deflating process by adding a new feature to the virtio-balloon device. And now, inflating the balloon to 3GB of a 4GB idle guest only takes 175ms, it's about 9 times as fast as before. TODO: optimize stage a by allocating/freeing a chunk of pages instead of a single page at a time. Signed-off-by: Liang Li <liang.z.li at intel.com> --- drivers/virtio/virtio_balloon.c | 199 ++++++++++++++++++++++++++++++++++-- include/uapi/li...

...of an idle guest with 4GB RAM. Should take 3 requests then, right? > Now: > total inflating time: 338ms > the count of virtio data transmission: 373 Why was this so high? I would expect 3 transmissions. > the call count of madvise: 865 > > before: > total inflating time: 175ms > the count of virtio data transmission: 1 > the call count of madvise: 42 > > Maybe the result will be worse if the guest is not idle, or the guest has more RAM. > Do you want more data? > > Is it worth to do that? > > Liang Either my math is wrong or there's an...

...* box is at my home and the other is in the office (before anyone > freaks this is a test environment)... Provided nothing is using the line at > the same time I am able to carry 1 voice call over the dialup link with no > real latency.. Ping times when no call is in session are about 165-175ms and > when a call is active 190-240ms.. I just ran a call for 5 mins to see if the > ping time climbed as you are experiencing and it didn't increase at all.. > > > > So you may have some other issue causing the increase in ping times.. > > > > Sorry I probably was...

...nd stage d can also be reduced a lot. > > This patch is the kernel side implementation which is intended to speed > up the inflating & deflating process by adding a new feature to the > virtio-balloon device. And now, inflating the balloon to 3GB of a 4GB > idle guest only takes 175ms, it's about 9 times as fast as before. > > TODO: optimize stage a by allocating/freeing a chunk of pages instead > of a single page at a time. Not commenting on the approach, but... > > Signed-off-by: Liang Li <liang.z.li at intel.com> > --- > drivers/virtio/virt...

Hi, I am using IAX to communicate between 2 sites, each site is using a 256k/64k ADSL Connection. I have noticed that when I connect my ping time to my 1st hop jumps from 30~ms to over 12,000ms in over a period of about 10 minutes, it just keeps climbing until the link is saturated. Naturally, there is a very long delay when speaking. What bandwidth would be adequate for IAX? or how can I

...e? for (pfn = min pfn; pfn < max pfn; pfn += 1G) { foreach page if page pfn < pfn || page pfn >= pfn + 1G continue set bit tell host } > > > > > > the call count of madvise: 865 > > > > > > before: > > > total inflating time: 175ms > > > the count of virtio data transmission: 1 the call count of madvise: 42 > > > > > > Maybe the result will be worse if the guest is not idle, or the guest has > > more RAM. > > > Do you want more data? > > > > > > Is it worth to do tha...

...educed a lot. > > > > This patch is the kernel side implementation which is intended to > > speed up the inflating & deflating process by adding a new feature to > > the virtio-balloon device. And now, inflating the balloon to 3GB of a > > 4GB idle guest only takes 175ms, it's about 9 times as fast as before. > > > > TODO: optimize stage a by allocating/freeing a chunk of pages instead > > of a single page at a time. > > > > Signed-off-by: Liang Li <liang.z.li at intel.com> > > --- > > drivers/virtio/virtio_ballo...

...educed a lot. > > > > This patch is the kernel side implementation which is intended to > > speed up the inflating & deflating process by adding a new feature to > > the virtio-balloon device. And now, inflating the balloon to 3GB of a > > 4GB idle guest only takes 175ms, it's about 9 times as fast as before. > > > > TODO: optimize stage a by allocating/freeing a chunk of pages instead > > of a single page at a time. > > > > Signed-off-by: Liang Li <liang.z.li at intel.com> > > --- > > drivers/virtio/virtio_ballo...

...nd stage d can also be reduced a lot. > > This patch is the kernel side implementation which is intended to speed > up the inflating & deflating process by adding a new feature to the > virtio-balloon device. And now, inflating the balloon to 3GB of a 4GB > idle guest only takes 175ms, it's about 9 times as fast as before. > > TODO: optimize stage a by allocating/freeing a chunk of pages instead > of a single page at a time. > > Signed-off-by: Liang Li <liang.z.li at intel.com> > --- > drivers/virtio/virtio_balloon.c | 199 ++++++++++++++++++...

...gt; > > > This patch is the kernel side implementation which is intended to > > > speed up the inflating & deflating process by adding a new feature to > > > the virtio-balloon device. And now, inflating the balloon to 3GB of a > > > 4GB idle guest only takes 175ms, it's about 9 times as fast as before. > > > > > > TODO: optimize stage a by allocating/freeing a chunk of pages instead > > > of a single page at a time. > > > > > > Signed-off-by: Liang Li <liang.z.li at intel.com> > > > --- > &gt...

...dom0 48%, target 68%, sent 3194Mb/s, dirtied 59Mb/s 18117 pages 2: sent 17470, skipped 632, delta 6237ms, dom0 37%, target 68%, sent 91Mb/s, dirtied 5Mb/s 1038 pages 3: sent 995, skipped 28, delta 211ms, dom0 32%, target 39%, sent 154Mb/s, dirtied 92Mb/s 594 pages 4: sent 546, skipped 27, delta 175ms, dom0 35%, target 83%, sent 102Mb/s, dirtied 9Mb/s 53 pages 5: sent 49, skipped 5, delta 48ms, dom0 8%, target 37%, sent 33Mb/s, dirtied 9Mb/s 14 pages 6: sent 9, skipped 5, Start last iteration094) Saving memory pages: iter 6 0% [2007-10-19 18:02:56 4489] DEBUG (__init__:1094) suspend [2007-10...