ssmoot

76 ms·

601.
by ssmoot 16y ago
Is this really standard practice for measuring the SLA? The contracts I've seen for a couple small businesses are generally per billable period. Which always made sense to me. I pay you for 99.5% uptime this month. If you don't achieve it,
602.
by ssmoot 16y ago
I forget the exact names, but go with it. :-) ZFS has two tables: A Block Pointer Table. This is your hashes of blocks and stays in RAM to let you know where on disk you'll find a particular block. It's also referenced during writing to ens
603.
by ssmoot 16y ago
It also takes a lot of RAM. 380 bytes for each de-dupe entry (block pointer) IIRC? I think it translated to something like 1GB of RAM per TB of storage assuming half of it was de-duped. It's really a useless feature when you do the math. Un
604.
by ssmoot 16y ago
I almost always use lzjb compression. It's nearly free and gives good returns for mixed data like the sort you'll find on system drives. Binaries, text, etc. Also the reliability aspect. I haven't had HFS+ go bad on my own systems, but I've
605.
by ssmoot 16y ago
I'm pretty sure the ZIL is not mirrored. If it were, it wouldn't increase performance since you'd still have to wait for the disks to fsync their copy to guarantee reliability. This is why Sun Storage products (and Nexenta, or homegrown clu
606.
by ssmoot 16y ago
Except you could do better-than-time-machine, cheap, continuous replication. You could even TimeMachine-without-additional-drives... I mean, why should my snapshot system and my backup system be one? How about letting me use TimeMachine loc
607.
by ssmoot 16y ago
Given the context, this seems a bit absurd.
608.
by ssmoot 16y ago
I've backed up a few hundred GB with Backblaze. I think the implication being made here is unfair. I'm much more likely to believe any issues with my upload were caused by shady AT&T QoS. You know the kind. Goto any speedtest site and s
609.
by ssmoot 16y ago
Most databases that I've worked with that need to stand up to any real load do end up getting fairly denormalized. So a database might contain multiple data silos. A job is a job and data is data. Attacking the interface doesn't seem entire
610.
by ssmoot 16y ago
Not that I've ever seen referenced. You don't really see the YARV name thrown a ton these days either I guess, but MRI means C-Ruby to most I think. YARV is a VM and from what I understand Matz had very little to do with it in the period le
611.
by ssmoot 16y ago
No offense, good comment. But the same is true for the Plumber, the Framer, the Project Manager, etc. To a Programmer, the Owner who pushes back on the customer is a good business person. To a Project Manager, the Owner who squeezes the pro
612.
by ssmoot 16y ago
It's faster than MRI. It allows you to write Java instead of C extensions. For developers without a C background, but with a Java or c# background that's pretty huge IMO. Plus instead of running 16 Unicorn workers for an app, eating up over
613.
by ssmoot 16y ago
No offense, but I'm going to be blunt: If you were putting yourself through college, paying as you go, then as long as you managed a low "B" I'll probably cut you a break. You'll get a fair shot. If you decide to skip college and throw your
614.
by ssmoot 16y ago
That first bit only works if you're already backing up to AWS. Moving terra-bytes of data into the cloud isn't quick and spinning up servers quickly isn't something that's going to help if your primary site goes down for many (most?) types
615.
by ssmoot 16y ago
As someone else stated, AWS doesn't provide you systems administrators. Storage administrators sure, but that's not difficult to absorb. So taking that out of the equation, you're left with hardware and bandwidth. I imagine Netflix didn't g
616.
by ssmoot 16y ago
No. It's much more cost effective to run your own infrastructure. It's always struck me as irresponsible for reedit to spend the sums they do for example. You could easily spend half that and get four times the hardware (in addition to an
617.
by ssmoot 16y ago
Nobody's dying here. There's a crazy amount of talk here about WL trying to bring down government or the collateral damage to business and equating that to civilian deaths. I don't know if Assange goes around saying "anarchy for everyone!"
618.
by ssmoot 16y ago
DM was nowhere near stable during the original development of ARel. O/R Mappers are difficult. The more ambitious, the more difficult, and DM is very ambitious. ;-) I can say it had absolutely nothing to do with ego. There was a actually a
619.
by ssmoot 16y ago
I don't know that it's really much to do with using the right tool for the job. Risk, reliability, performance expectations, etc. I don't know if I've ever met a project of any significant complexity that would have been more of a success i
620.
by ssmoot 16y ago
Ditto. Though I'd narrow "cross platform" to non-Microsoft. No hate for MS, but I don't see deploying Windows servers and associated licensing costs any more likely for me than buying Oracle licenses. So yeah, as a small Ruby shop that was
621.
by ssmoot 16y ago
How much is storage? zfs has some utility for small filesystems (less than 1TB) for backups and reliability, but it's real strength is in storage consolidation. The problem with cloud storage so far is that it can easily cost 10X the price
622.
by ssmoot 16y ago
I really don't get why people feel the need to reinvent row/column structures when there's a more reliable, better option built into every browser for the past decade. "Tables are for tabular data" just strikes me as ego, and while I've see
623.
by ssmoot 16y ago
Minor correction, Scrub just rewrites all data. And since ZFS is copy-on-write and compares new blocks to old, you'll find corruption or bad blocks that way. Also, since ZFS doesn't dedupe (just don't use it) or compress at rest, a scrub is
624.
by ssmoot 16y ago
I've heard about btrfs being "just around the corner" for a good while now. ZFS on Nexenta and even FreeBSD is pretty robust. It's a great storage platform. I wouldn't let waiting on a stable btrfs make my SAN/NAS decisions for me. The othe
625.
by ssmoot 16y ago
So I'd guess EBS is nothing more than ISCSI targets then. ZFS won't do anything to make those usable on multiple machines at once. ZFS does make it easy to mirror multiple LUNs for redundancy though, so if you have a timeout or permanent di
626.
by ssmoot 16y ago
If it's as small as a single server, you probably don't need an SSD to cache access. If your db server is partitioned out, just put the whole fs on SSD. You can't partition the L2ARC, so you really don't want to mix web assets with database
627.
by ssmoot 16y ago
ZVOLs are block devices. So you could install another fs on top of one. Or more usefully export it as an iSCSI target.
628.
by ssmoot 16y ago
I used to know OCZ as a bit of a shady, benchmark gaming outfit "back in the day". Today however, the SandForce based Vertex II Pro SSDs are amazing. 10K+ IOPs sustained with a genuine database load. 100GB for just over $600. Compare that w
629.
by ssmoot 16y ago
Running VMs for testing during development (ie: an IE6 VM, IE7 VM, etc) are the major driver for us. Outside of that, my work machine has Chrome, Mail.app, Terminal and Textmate open generally so 2GB is fine without the VMs. My home machine
630.
by ssmoot 16y ago
"It needs to at least compete with Java/PHP ect." Why? And you know what? It never will. Rubyists need to give this up. It's a pipe-dream. Microsoft can set up large/local conferences, charge $75, and make a sales pitch. Dozens of vendors l

More ›