[openib-general] Getting rid of pinned memory requirement
Michael Krause
krause at cup.hp.com
Tue Mar 15 09:51:07 PST 2005
At 05:35 PM 3/14/2005, Caitlin Bestler wrote:
>
>
> > -----Original Message-----
> > From: Troy Benjegerdes [mailto:hozer at hozed.org]
> > Sent: Monday, March 14, 2005 5:06 PM
> > To: Caitlin Bestler
> > Cc: openib-general at openib.org
> > Subject: Re: [openib-general] Getting rid of pinned memory requirement
> >
> > >
> > > The key is that the entire operation either has to be fast
> > > enough so that no connection or application session layer
> > > time-outs occur, or an end-to-end agreement to suspend the
> > > connetion is a requirement. The first option seems more
> > > plausible to me, the second essentially
> > > reuqires extending the CM protocol. That's a tall order even for
> > > InfiniBand, and it's even worse for iWARP where the CM
> > > functionality typically ends when the connection is established.
> >
> > I'll buy the good network design argument.
I and others designed InfiniBand RNR (Receiver not ready) operations to
allow one to adjust V-to-P mappings (not change the address that was
advertised) in order to allow an OS to safely play some games with memory
and not drop a connection. The time values associated with RNR allow a
solution to tolerate up to infinite amount of time to perform such
operations but the envisioned goal was to do this on the order of a handful
or milliseconds in the worse case. For iWARP, there was no support for
defining RNR functionality as indeed many people claimed one could just
drop in-bound segments and allow the retransmission protocol to deal with
the delay (even if this has performance implications due to back-off
algorithms though some claim SACK would minimize this to a large
extent). Again, the idea was to minimize the worse case to milliseconds of
down time. BTW, all of this assumed that the OS would not perform these
types of changes that often so the long-term impact on an application would
be minimum.
> >
> > I suppose if the kernel wants to revoke a card's pinned
> > memory, we should be able to guarantee that it gets new
> > pinned memory within a bounded time. What sort of timing do
> > we need? Milliseconds?
> > Microseconds?
> >
> > In the case of iWarp, isn't this just TCP underneath? If so,
> > can't we just drop any packets in the pipe on the floor and
> > let them get retransmitted? (I suppose the same argument goes
> > for infiniband..
> > what sort of a time window do we have for retransmission?)
> >
> > What are the limits on end-to-end flow control in IB and iWarp?
> >
>
> >From the RDMA Provider's perspective, the short answer is "quick enough
> so that I don't have to do anything heroic to keep the connection alive."
It should not require anything heroic. What is does require is a local
method to suspend the local QP(s) so that it cannot place or read memory in
the effected area. That can take some time depending upon the
implementation. There is then the time to over write the mappings which
again depending upon the implementation and the number of mappings could be
milliseconds in length.
>With TCP you also have to add "and healthy". If you've ever had a long
>download that got effectively stalled by a burst of noise and you just hit
>the 'reload' button on your browser then you know what I'm talking about.
>
>But in transport neutral terms I would think that one RTT is definitely
>safe -- that much data could have
>been dropped by one switch failure or one nasty spike in inbound noise.
>
> > >
> > > Yes, there are limits on how much memory you can mlock, or even
> > > allocate. Applications are required to reqister memory precisely
> > > because the required guarantess are not there by default.
> > Eliminating
> > > those guarantees *is* effectively rewriting every RDMA application
> > > without even letting them know.
> >
> > Some of this argument is a policy issue, which I would argue
> > shouldn't be hard-coded in the code or in the network hardware.
> >
> > At least in my view, the guarantees are only there to make
> > applications go fast. We are getting low latency and high
> > performance with infiniband by making memory registration go
> > really really slow. If, to make big HPC simulation
> > applications work, we wind up doing memcpy() to put the data
> > into a registered buffer because we can't register half of
> > physical memory, the application isn't going very fast.
> >
>
>What you are looking for is a distinction between registering
>memory to *enable* the RNIC to optimize local access and
>registering memory to enable its being advertised to the
>remote end.
>
>Early implementations of RDMA, both IB and iWARP, have not
>distinquished between the two. But theoretically *applications*
>do not need memory regions that are not enabled for remote
>access to be pinned. That is an RNIC requirement that could
>evolve. But applications themselves *do* need remotely
>accessible memory regions, portions of which they intend
>to advertise with RKeys, to be truly available (i.e., pinned).
>
>You are also making a policy assumption that an application
>that actually needs half of physical memory should be using
>paged memory. Memory is cheap, and if performance is critical
>why should this memory be swapped out to disk?
>
>Is the limitation on not being able to register half of
>physical memory based upon some assumption that swapping
>is a requirement? Or is it a limitation in the memory region
>size? If it's the latter, you need to get the OS to support
>larger page sizes.
For some OS, you can pin very large areas. I've seen 15/16 of memory being
able to be pinned with no adverse impacts on the applications. For these
OS, kernel memory is effectively pinned memory. As such, depending upon
the mix of services being provided, the system may operate quite nicely
with such large amounts of memory being pinned. As more services are
"ported" to operate over RDMA technologies, memory management isn't
necessarily any harder; it just becomes something people have to think more
about. Today's VM designs have allowed people to get sloppy as they assume
that swapping will occur and since many platforms are not that loaded, they
don't see any real adverse impacts. User-space RDMA applications requires
people to think once again about memory management and that swapping isn't
a get-out-of-jail card. One needs to develop resource management tools to
determine who obtains specified amounts of resources and their
priorities. For the most part, this is somewhat a re-invention of some
thinking that went into the micro-kernel work in past years. These
problems are not intractable; they are only constrained by the legacy
inertia inherent in all technologies today.
Mike
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openfabrics.org/pipermail/general/attachments/20050315/9502e11c/attachment.html>
More information about the general
mailing list