And the tile processing is to merge the existing map tiles with newly updated ones. I don't know how they do this, but at some point, the data does need to be merged into the exact same system the routing server and other realtime components live on.

tibble wrote:You may be missing the fact that the speeds gathered from all the users are saved to the db 24/7.
This means once the snapshot db is updated with new tiles they'd then have to merge the changes from the 'production' DB.

Yes and no. The realtime segment speed (traffic layer) is part of Live, however historical data is part of the edit database and comes from drive analysis. You can see this for yourself on any newly created segment, it will start to collect average speed data from every drive across it without being part of the Live/Routing map. So basically xteejx is still right, there's no logical reason for them to be tied together, they shouldn't even be on the same database server or datastore.

I agree with gettingthere. Sure AWS is great and scalable, but unless the software can still function correctly in a scalable, load balanced environment, it doesn't matter how great your cloud service is. I've worked on plenty of applications that just flat out don't work in load balanced environments due to poor or short-sighted design.

The good news is that this is all fixable, it just takes time, something most people aren't willing to give anymore.

Often wme is offline for some time so it's not like something new for us. I think the idea has some merit.

David using Tapatalk

The only thing that surprises me is that you're surprised!

via Tapatalk

@gettingthere, yep, I totally agree. It's the same with every version refresh though with two differences, bigger user base to annoy (and greater attrition) and competition is more advanced!

via Tapatalk

Yet another good reason to not map every parking lot on the planet.

They need to sort out their bank account. We all know they use AWS. It's scalable to absolute immense power if you pay for it. Look it up, elastic load balancing, seriously there is NO excuse anymore for these problems. I said it before, if you want to support more users, bring more servers online, it's what Amazon do and it's why people pay to use their services... Scalability and parallel processing... I was shrugged off during the Paris meet up (didn't attend dmcconachie spoke on my behalf about it) about these issues, and we were all told it was software issues, not processing power. Well I now know as well as you guys that that's just to fob us off. Don't get me wrong I love Waze and want to see it make it, but unless truths are told and we're not treated as if we're stupid it's just not going to work.
To have respect, you need to respect your community. Infrastructure is what makes any of the system even possible. Get that fixed so that it really IS scalable and release as much as you like. Until that time comes, it will just be one problem after another until everyone gets fed up with it and leaves, which I'm sure will be heartbreaking for most of us that have devoted our time and resources on this product.

Sent from my GT-I9100P using Tapatalk 2

Exactly....
So why push stupid poxy client updates out at all? Rewrite the server software and have it done.
To be honest, tile gen shouldn't even be anywhere NEAR the real time stuff.
What's wrong with zsyncing a snapshot of the database and generating tiles on another server? It's pretty damn obvious it's processor and memory intensive. It should be completely separate.
Really, the issue here is blatent disregard for the obvious.

Sent from my GT-I9100P using Tapatalk 2