Why Dropbox decided to drop AWS and build its own infrastructure and network

Posted by rbanffy on Friday September 15, 2017 @10:18AM
from the techcrunch.com dept.

https://techcrunch.com/2017/09/15/why-dropbox-decided-to-drop-aws-and-build-its-own-infrastructure-and-network/

( 9 comments | 53 points | on HN )

Re:Why Dropbox decided to drop AWS and build its own infrast... by kylehotchkiss on Friday September 15, 2017 @01:17PM (#15258884)

I recently switched from Dropbox to Google Drive (unlimited storage is nice) to consolidate all my cloud files (had an epiphany that iCloud Drive, Dropbox, Google Drive, and AWS were just too many cloud storages to pay for) and I miss dropbox's sync tool already. Was a nice quality product and I will miss it.

[ Parent | on HN ]

Re:Why Dropbox decided to drop AWS and build its own infrast... by jgowdy on Friday September 15, 2017 @08:49PM (#15262081)

I don't trust "unlimited" storage anymore. From anyone. Having had to evacuate my data from Microsoft OneDrive, Amazon Drive, and CrashPlan, I'm done with that. At least when I'm using Backblaze B2 I know I'm paying for the usage as is everyone else, so pirates aren't going to screw up the cost model and have me doing another massive data migration.

[ Parent | on HN ]

Re:Why Dropbox decided to drop AWS and build its own infrast... by sidcool on Friday September 15, 2017 @02:42PM (#15259593)

I have heard this several times, people really like Dropbox sync. What I cannot understand is what prevents Google from making a kick-ass sync client?

[ Parent | on HN ]

Re:Why Dropbox decided to drop AWS and build its own infrast... by smmnyc on Friday September 15, 2017 @03:38PM (#15260108)

Google is going to split their sync tool into a consumer tool (Backup and Sync) and business tool (Drive File Stream); I think they stopped updating their main sync client a while back in preparation for this split. I think the business tool is the only one that will allow for files to appear locally that aren't actually downloaded. It's still in beta.

[ Parent | on HN ]

Re:Why Dropbox decided to drop AWS and build its own infrast... by kylehotchkiss on Friday September 15, 2017 @05:36PM (#15260973)

I am actually talking about quality of Backup and Sync. It's got about 90gb to upload and it's using 130% of processor and clicking it hardly shows progress because it's so bogged. Hopefully they fix soon, otherwise the selective sync is eons better than dropbox's implementation.

[ Parent | on HN ]

Re:Why Dropbox decided to drop AWS and build its own infrast... by noncoml on Friday September 15, 2017 @07:45PM (#15261772)

The article is very vague with not enough technical or business details about the move and the reasons and challenges behind it.

[ Parent | on HN ]

Re:Why Dropbox decided to drop AWS and build its own infrast... by paxy on Friday September 15, 2017 @08:38PM (#15262034)

This one (from March 2016) is way better in every way. Not sure why this TechCrunch article even needed to be written now.

https://www.wired.com/2016/03/epic-story-dropboxs-exodus-ama...

[ Parent | on HN ]

Re:Why Dropbox decided to drop AWS and build its own infrast... by MR4D on Saturday September 16, 2017 @07:15PM (#15267277)

Revenue. TechCrunch doesn't make revenue off of Wired's articles.

[ Parent | on HN ]

Re:Why Dropbox decided to drop AWS and build its own infrast... by jgowdy on Friday September 15, 2017 @08:46PM (#15262073)

My bet is they can't be competitive with Backblaze B2 when they're on top of S3. Every S3 based cloud storage provider is going to be S3 special negotiated price plus minimal profit. The pricing of B2 is almost unmatched for random access storage at half a cent a gig. That's literally half of Dropbox's 1TB price. I've been backing up to Backblaze B2 via Arq on OSX, so I can manage my own archives and not have things deleted from backup when I delete them from disk.

[ Parent | on HN ]