Re: Speeding up slooooowwww extractions

[Scott Wheeler <scott@directededge.com>]

> On 26 May 2021, at 23:02, Graham Percival <gperciva@tarsnap.com> wrote:
> 
> On Wed, May 26, 2021 at 10:35:37PM +0200, Scott Wheeler wrote:
>> 
>>> On 26 May 2021, at 22:19, hvjunk <hvjunk@gmail.com> wrote:
>>> 
>>> --fast-read also not the next issue… but yeah, looks like that latency is the killer for my use case ;(
>> 
>> I don’t think latency is the issue:
> 
> Colin specifically says that latency is the problem:

My read of that is that there’s an algorithm written for local memory that is executing on network data.  Even on the same local network, there are several orders of magnitude in the latency of those classes.

When I’ve needed to pull huge amounts of data out of tarsnap, I’ve done it from the same AWS Region as tarsnap.  It’s still very slow.  (Or was ages ago when I was testing.)

-Scott