Okay. Thank you. Would a specific database be similar enough where transition later can be much easi
Okay. Thank you. Would a specific database be similar enough where transition later can be much easier? Would planet scale be the next best option?
JI didn't put a lot of effort into research. I just knew that planetscale would meet my needs and had a serverless driver.
you might also want to check out:
• Turso https://chiselstrike.com
• xata.io
you might also want to check out:
• Turso https://chiselstrike.com
• xata.io
Eliminate application bottlenecks by bringing the performance and data portability of open source, lightweight SQL close to your users with low overhead
SThis is working good
M@elithrar ^
MWe’ll be announcing (just announcing: not enabling) pricing in a few weeks, but it’s very much an “it depends” case. DO is key-value based, whereas D1 is a queryable database. If you have a lot of small reads, then D1 is likely to be much cheaper. DO charges on 4KB boundaries. I’d need to know a LOT more about your query patterns here.
MWhat are the limits of d1
MLike as far as how much data it can store
M(Remembering that D1 is an alpha: a true alpha, not a marketing alpha!) 
MSo if d1 is 100mb limit I’m trying to understand how that can be useful? Like say I’m storing e-commerce orders. What do we do when we reach a limit
MIt’s an alpha. Per-DB storage limits will increase dramatically when we get to beta. Hold tight!
MRegardless though just trying to understand how that would be handled in prod. Say it’s 10gb even. What do I do when I reach my cap
MDO prices a read & write on 4KB boundaries. Many web-facing database workloads are selecting single rows of user data (500 bytes!), or simple reads to generate lists. These are very small.
MWe’ll the thing is a database is forever expanding right. If you are constantly adding data to your db you are guaranteed to one day reach that cap. Just trying to understand how one would handle that
MSame thing if you hit 50GB? It takes a lot of traditional relational data to add up, and you will be able to create many databases. Each DB will have its own limits.
MSo we would just create a new database to store the orders after the first one adds up you’re saying?
MSorry if I’m not understanding this correctly as I have mostly worked with DO. Like planet scale I’m pretty sure if you reach your plan limit you just upgrade and get more storage. So with d1 I’m trying to understand what efforts will be required
MI’m building a e-commerce saas. The order data will add up very quickly
MI think it’s worth stepping back a second. I understand that hitting a storage limit sounds concerning, but… if you have 500 bytes per row, and 10M order rows: that’s just 5GB.
MSo let’s say I reach that In a year with 100 tenants
EFrom https://developers.cloudflare.com/d1/
While in the Alpha period, there is a possibility of breaking changes. The Alpha is meant for testing purposes and using it for Production traffic is not recommended.
MI’m just going to restate here: D1 is in alpha. A lot will change. I’d love to share more, but much of this is subject to change. Storage limits included.
MYeah my concern is not that it’s in alpha and only 100mb but rather if we can’t expand storage what do we do when we reach limits. Sorry if I’m going in circles I’m new to data storage and just trying to learn. I love CF and a fanboy of all services. Trying to do everything within the ecosystem
MAs with any DB - you end up potentially splitting/sharding it out - by date or user ID or order ID. Even if you can scale AWS RDS up to 1TB, you often don’t want to do that as it ultimately has performance and risk trade-offs. Same with PlanetScale.
MWhat about with something like planet scale?
MI’m just going to reiterate the need to hold tight: I haven’t said anything concrete about per-DB limits, and we’re not quite ready to share those.
MOkay. Tysm Very excited.
Kmy project is also in alpha, so I'll go beta once D1 goes into beta lol
AHi @elithrar, can you give us a steer around how we can expect to handle breaking changes? Is there going to be a compatibility flag/date or will it just stop working?
MCloudflare
Cloudflare Workers KV is low-latency key-value storage at the edge. Leverage serverless computing and build serverless architecture with key-value storage.
UOr #durable-objects
DHello, I'm confused as to how im supposed to upload sql files to d1.
When using wrangler d1 excecute <db> --file=file.sql
i get this error:
When using wrangler d1 excecute <db> --file=file.sql
i get this error:
KYou've typo'd
executeokay this never happened
Dthank you
Mif you want to try something super bleeding edge, you can also paste sql files into the console:
Dquery is too long for that 
MI've pasted a 500KB query in there, 2200 rows inserted no worries
Mmight lag tho
D140k here
Dgood news, the console seems to be better than wrangler. Inserted 140k rows in about 15s no problem.
Thank you for the suggestion, props to whoever worked on this feature its amazing
Thank you for the suggestion, props to whoever worked on this feature its amazing
JIs there a timeline of when things are announced like pricing announced with beta. not like eta but what features will be shipped