Some call it signs and portents, others good fortune, but a lot of people are starting to show an interest in object persistence. At least we think they are, but that might just be a symptom of obsessive-compulsive disorder. But anyway.
- From Virtual Memory To Object Storage
- BYTE's Jon Udell on object persistence as an elegant solution to
the current need for virtual memory:
"What would an OS would look like that was built from the ground up in an OO language whose data objects automatically persisted to an object database? The OODB might even be in cahoots with OO language's garbage collector."
- Trees Are Fast
- Hans Reiser's seminal whitepaper on optimising filesystem
performance, as demonstrated by his own reiserfs.
- EPCKPT
- A Linux kernel modification for checkpoint/restarting running processes. We anxiously await its wholesale adoption.
- Persistent Object Store Applications
- A good (although now almost antique) overview of the issues arising from persistence. The abstract says: This thesis seeks to advise application programmers on the costs and benefits of programming in persistent environments.
Post++ is a good practical approach to persistence. It requires that you modify your code more extensively than ColdStore does.
Though it sometimes feels that way, we're not the only ones thinking
about these things. A few references to articles on object persistence
and all that gubbins.
Why ColdStore? An explanation of what we're up to.
The nitty-gritty: an explanation of ColdStore's design, what it
does, and what it might be used for.
Get it, compile it, run it, tweak it. Lather, rinse,
repeat: all with the fresh smell of GNU.
We think that ColdStore has potential. Here's what's in the
offing: from the nearly feasible to the bright blue sky. You can
probably help us out here.
Praise the visionaries behind this thing; alternatively,
berate the guilty parties.
That Oscar acceptance speech in full.
