Hardware for EROS?

[Valerio Bellizzomi]

Hello,
to achieve high redundancy on volatile memories we could use also
"reflective
memory" which is an architecture with many memory boards connected together
by a bus constructed with optical physical links between adjacent boards,
every time a write occurs on a board the change is reflected to all others.
Each board can read data back from all others.
However, I don't actually see if there can be any useful application of
this
technology to the EROS domain, note also that there is no backup battery
on boards, but ultimately, to take an example of what I am talking about
you
can look at http://reflectivememory.vmic.com

Sincerely,
val.

*********** REPLY SEPARATOR  ***********

On 11/04/2000 at 13.42 Bill Frantz wrote:

>Many people have direct experience with battery backuped up main memory
>permanent storage, because that is what the Palm Pilot uses.  Those
>experiences are directly applicable to larger systems.
>
>Like most things, it works most of the time, but sometimes you need to go
>to another level of backup.  You can certainly postulate failures that
will
>kill all your backups.  The sun going nova is my favorite example.
>
>At 03:13 PM 4/11/00 -0400, Jonathan S. Shapiro wrote:
>>> If a disk seek is the cost of a commit, then many realistic distributed
>>> computing scenarios cannot wait that long before releasing a message.
>>
>>Typical seek delay will be lower on a system using EROS-style
checkpoints,
>>because the arm is already in the checkpoint zone with high likelihood.
It's
>>still units of milliseconds, though.
>>
>>> It would seem that a vanilla EROS system
>>> + a UPS sufficient to keep things going until the next two commits (ie,
>>> until the next checkpoint is stable) could validly be considered an
EROS
>>> that never needed to roll back because of power outages.
>>
>>Better still, you can detect remaining power and just start checkpointing
>>more frequently. The journaled stuff never rolls back anyway.
>>
>>> Does EROS check invariants before taking a checkpoint?
>>
>>Yes. If the invariant check fails the system reboots. At the moment it
>>doesn't know how to warm-boot yet, but I hope to fix that.  Reboot means
>>that it recovers from the most recent stable checkpoint.
>>
>>I think in practice that the checkpoint really has to be stabilized to
>>non-volatile storage. If there is any practical chance that the battery
dies
>>on you, the entire system ends up unrecoverable unless the checkpoint has
>>also been stabilized.
>>
>>shap
>>
>>
>>
>>