I was previously building accounting software for firms with millions of transactions. I desperately needed to move from Postgres to an OLAP database but didn\u2019t know where to start. I eventually built abstractions around Clickhouse: My application code called an insert() function but in the background I had to stand up Kafka for streaming, bulk loading, DB drivers, Clickhouse configs, and manage schema changes.
This was all a big distraction when all I wanted was to save data and get it back. So I decided to build a better developer experience around it. The software is open-source: https://github.com/scratchdata/ScratchDB and and the paid offering is a hosted version: https://www.scratchdb.com/.
It's called \u201cScratchDB\u201d because the idea is to make it easy to get started from scratch. It\u2019s a massively simpler abstraction on top of Clickhouse.
ScratchDB provides two endpoints [1]: one to insert data and another to query. When you send any JSON, it automatically creates tables and columns based on the structure [2]. Because table creation is automated, you can just start sending data and the system will just work [3]. It also means you can use Scratch as any webhook destination without prior setup [4,5]. When you query, just pass SQL as a query param and it returns JSON.
It handles streaming and bulk loading data. When data is inserted, I append it to a file on disk, which is then bulk loaded into Clickhouse. The overall goal is for the platform to automatically handle managing shards and replicas.
The whole thing runs on regular servers. Hetzner has become our cloud of choice, along with Backblaze B2 and SQS. It is written in Go. From an architecture perspective I try to keep things simple - want folks to make economical use of their servers.
So far ScratchDB has ingested about 2 TB of data and 4,000 requests/second on about $100 worth of monthly server costs.
Feel free to download it and play around - if you\u2019re interested in this stuff then I\u2019d love to chat! Really looking for feedback on what is hard about analytical databases and what would make the developer experience easier!
[1] https://scratchdb.com/docs
[2] https://scratchdb.com/blog/flatten-json/
[3] https://scratchdb.com/blog/scratchdb-email-signups/
[4] https://scratchdb.com/blog/stripe-data-ingest/
[5] https://scratchdb.com/blog/shopify-data-ingest/"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Show HN: ScratchDB \u2013 Open-Source Snowflake on ClickHouse"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/scratchdata/ScratchDB"}},"_tags":["story","author_memset","story_38038239","show_hn"],"author":"memset","children":[38038468,38041892,38041920,38042000,38042460,38042522,38042817,38042892,38042978,38043348,38045020,38045290,38046421,38047489,38051282,38052448],"created_at":"2023-10-27T13:34:06Z","created_at_i":1698413646,"num_comments":56,"objectID":"38038239","points":261,"story_id":38038239,"story_text":"Hello! For the past year I\u2019ve been working on a fully-managed data warehouse built on Clickhouse. I built this because I was frustrated with how much work was required to run an OLAP database in prod: re-writing my app to do batch inserts, managing clusters and needing to look up special CREATE TABLE syntax every time I made a change. I found pricing for other warehouses confusing (what is a \u201ccredit\u201d exactly?) and worried about getting capacity-planning wrong.
I was previously building accounting software for firms with millions of transactions. I desperately needed to move from Postgres to an OLAP database but didn\u2019t know where to start. I eventually built abstractions around Clickhouse: My application code called an insert() function but in the background I had to stand up Kafka for streaming, bulk loading, DB drivers, Clickhouse configs, and manage schema changes.
This was all a big distraction when all I wanted was to save data and get it back. So I decided to build a better developer experience around it. The software is open-source: https://github.com/scratchdata/ScratchDB and and the paid offering is a hosted version: https://www.scratchdb.com/.
It's called \u201cScratchDB\u201d because the idea is to make it easy to get started from scratch. It\u2019s a massively simpler abstraction on top of Clickhouse.
ScratchDB provides two endpoints [1]: one to insert data and another to query. When you send any JSON, it automatically creates tables and columns based on the structure [2]. Because table creation is automated, you can just start sending data and the system will just work [3]. It also means you can use Scratch as any webhook destination without prior setup [4,5]. When you query, just pass SQL as a query param and it returns JSON.
It handles streaming and bulk loading data. When data is inserted, I append it to a file on disk, which is then bulk loaded into Clickhouse. The overall goal is for the platform to automatically handle managing shards and replicas.
The whole thing runs on regular servers. Hetzner has become our cloud of choice, along with Backblaze B2 and SQS. It is written in Go. From an architecture perspective I try to keep things simple - want folks to make economical use of their servers.
So far ScratchDB has ingested about 2 TB of data and 4,000 requests/second on about $100 worth of monthly server costs.
Feel free to download it and play around - if you\u2019re interested in this stuff then I\u2019d love to chat! Really looking for feedback on what is hard about analytical databases and what would make the developer experience easier!
[1] https://scratchdb.com/docs
[2] https://scratchdb.com/blog/flatten-json/
[3] https://scratchdb.com/blog/scratchdb-email-signups/
[4] https://scratchdb.com/blog/stripe-data-ingest/
[5] https://scratchdb.com/blog/shopify-data-ingest/","title":"Show HN: ScratchDB \u2013 Open-Source Snowflake on ClickHouse","updated_at":"2025-01-05T08:08:17Z","url":"https://github.com/scratchdata/ScratchDB"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"malisper"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Why is Snowflake so Valuable?"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://www.freshpaint.io/blog/why-is-snowflake-so-valuable"}},"_tags":["story","author_malisper","story_24641481"],"author":"malisper","children":[24641808,24641838,24641869,24641928,24641949,24641956,24641977,24642057,24642078,24642089,24642219,24642307,24642370,24642381,24642519,24642612,24642644,24642875,24642913,24643170,24644179,24646182,24646317,24646320,24646485,24646841,24646847,24647809,24647982,24648565,24648663,24649713,24662561],"created_at":"2020-09-30T17:50:53Z","created_at_i":1601488253,"num_comments":163,"objectID":"24641481","points":247,"story_id":24641481,"title":"Why is Snowflake so Valuable?","updated_at":"2025-10-06T21:08:26Z","url":"https://www.freshpaint.io/blog/why-is-snowflake-so-valuable"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"shalanah"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Hi all! Just released Super Snowflake Maker!
Draw on the pie with freeform or polygon tools, change the number of sections, click on the large snowflake to see fold, and.... download!
Enjoy + Happy Holidays!
(tech: threejs/r3f, react, ts, useSpring, tailwind, canvas, svg, offscreen canvas, paperjs)"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Show HN: Super Snowflake Maker"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://supersnowflakemaker.com"}},"_tags":["story","author_shalanah","story_42493832","show_hn"],"author":"shalanah","children":[42512117,42512209,42512389,42512488,42512507,42512548,42512612,42512810,42512982,42513049,42513486,42514568,42515847,42515886,42515961,42516683,42517317,42517325,42522919,42525877],"created_at":"2024-12-23T12:09:28Z","created_at_i":1734955768,"num_comments":48,"objectID":"42493832","points":228,"story_id":42493832,"story_text":"Hi all! Just released Super Snowflake Maker!
Draw on the pie with freeform or polygon tools, change the number of sections, click on the large snowflake to see fold, and.... download!
Enjoy + Happy Holidays!
(tech: threejs/r3f, react, ts, useSpring, tailwind, canvas, svg, offscreen canvas, paperjs)","title":"Show HN: Super Snowflake Maker","updated_at":"2026-02-28T10:12:12Z","url":"https://supersnowflakemaker.com"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"rxin"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Databricks response to Snowflake's accusation of lacking integrity"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://databricks.com/blog/2021/11/15/snowflake-claims-similar-price-performance-to-databricks-but-not-so-fast.html"}},"_tags":["story","author_rxin","story_29232346"],"author":"rxin","children":[29232654,29232655,29232688,29232698,29232720,29232723,29232743,29232761,29232778,29232805,29232821,29232834,29232937,29232972,29233047,29233077,29233096,29233100,29233184,29233217,29233547,29234035,29234493,29234558,29234986,29235454,29236331,29237762],"created_at":"2021-11-15T20:35:14Z","created_at_i":1637008514,"num_comments":156,"objectID":"29232346","points":217,"story_id":29232346,"title":"Databricks response to Snowflake's accusation of lacking integrity","updated_at":"2025-07-10T13:15:23Z","url":"https://databricks.com/blog/2021/11/15/snowflake-claims-similar-price-performance-to-databricks-but-not-so-fast.html"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"tkz1312"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"SnowflakeOS: Beginner friendly and GUI focused NixOS variant"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://snowflakeos.org/"}},"_tags":["story","author_tkz1312","story_41124472"],"author":"tkz1312","children":[41124974,41125009,41125041,41125210,41125234,41125365,41125418,41125653,41125935,41125951,41126052,41126117,41126452,41127436,41127689,41128646],"created_at":"2024-07-31T23:06:11Z","created_at_i":1722467171,"num_comments":166,"objectID":"41124472","points":215,"story_id":41124472,"title":"SnowflakeOS: Beginner friendly and GUI focused NixOS variant","updated_at":"2026-07-01T19:50:02Z","url":"https://snowflakeos.org/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"doomrobo"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Tor Snowflake Proxy"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://snowflake.torproject.org/"}},"_tags":["story","author_doomrobo","story_29634636"],"author":"doomrobo","children":[29635073,29635358,29635664,29635710,29635784,29635989,29636111,29636957,29637347,29637582,29640233],"created_at":"2021-12-21T06:32:11Z","created_at_i":1640068331,"num_comments":84,"objectID":"29634636","points":197,"story_id":29634636,"title":"Tor Snowflake Proxy","updated_at":"2025-12-14T20:35:08Z","url":"https://snowflake.torproject.org/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"whoiskatrin"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Most companies do not need Snowflake or Databricks"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://kjhealey.medium.com/cached-takes-80-of-companies-do-not-need-snowflake-or-databricks-5ebda64c0853"}},"_tags":["story","author_whoiskatrin","story_37207329"],"author":"whoiskatrin","children":[37207651,37207666,37207687,37207690,37207699,37207706,37207733,37207738,37207760,37207769,37207776,37207786,37207817,37207826,37207828,37208007,37208027,37208077,37208159,37208190,37208294,37208540,37208648,37208786,37208794,37209046,37209353,37209545,37209573,37209735,37209752,37209789,37209884,37210222,37210530,37210880,37210921,37210931,37211229,37213123,37213292,37233502],"created_at":"2023-08-21T09:55:14Z","created_at_i":1692611714,"num_comments":133,"objectID":"37207329","points":179,"story_id":37207329,"title":"Most companies do not need Snowflake or Databricks","updated_at":"2025-10-06T20:39:00Z","url":"https://kjhealey.medium.com/cached-takes-80-of-companies-do-not-need-snowflake-or-databricks-5ebda64c0853"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"kressaty"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Snowflake S-1"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://www.sec.gov/Archives/edgar/data/1640147/000162828020013010/snowflakes-1.htm"}},"_tags":["story","author_kressaty","story_24265041"],"author":"kressaty","children":[24265224,24265226,24265319,24265349,24265370,24265389,24265404,24265416,24265433,24265444,24265458,24265482,24265484,24265493,24265497,24265559,24265572,24265668,24265819,24265836,24276403,24276440,24276455,24298088],"created_at":"2020-08-24T20:18:35Z","created_at_i":1598300315,"num_comments":100,"objectID":"24265041","points":170,"story_id":24265041,"title":"Snowflake S-1","updated_at":"2024-09-20T06:49:19Z","url":"https://www.sec.gov/Archives/edgar/data/1640147/000162828020013010/snowflakes-1.htm"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"danielcampos93"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Neeva acquired by Snowflake"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://www.snowflake.com/blog/snowflake-acquires-neeva-to-accelerate-search-in-the-data-cloud-through-generative-ai/"}},"_tags":["story","author_danielcampos93","story_36063346"],"author":"danielcampos93","children":[36063993,36064316,36064341,36064443,36064488,36064689,36064703,36064717,36064941,36065167,36065885,36066042,36067347,36067369,36067538,36071136],"created_at":"2023-05-24T20:17:07Z","created_at_i":1684959427,"num_comments":111,"objectID":"36063346","points":166,"story_id":36063346,"title":"Neeva acquired by Snowflake","updated_at":"2026-05-12T19:53:55Z","url":"https://www.snowflake.com/blog/snowflake-acquires-neeva-to-accelerate-search-in-the-data-cloud-through-generative-ai/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"thomasdziedzic"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Building a personal data warehouse in Snowflake for fun and no profit"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://www.thomasdziedzic0.com/blog/building-a-personal-data-warehouse-in-snowflake-for-fun-and-no-profit"}},"_tags":["story","author_thomasdziedzic","story_25978000"],"author":"thomasdziedzic","children":[25978650,25978800,25978806,25979490,25979674,25979980,25980505,25980998,25981317,25982035,25982112,25982284,25982619,25983173,25986777],"created_at":"2021-01-31T11:24:09Z","created_at_i":1612092249,"num_comments":31,"objectID":"25978000","points":166,"story_id":25978000,"title":"Building a personal data warehouse in Snowflake for fun and no profit","updated_at":"2024-09-20T07:53:11Z","url":"https://www.thomasdziedzic0.com/blog/building-a-personal-data-warehouse-in-snowflake-for-fun-and-no-profit"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"mfiguiere"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Snowflake to buy Crunchy Data for $250M"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://www.wsj.com/articles/snowflake-to-buy-crunchy-data-for-250-million-233543ab"}},"_tags":["story","author_mfiguiere","story_44162443"],"author":"mfiguiere","children":[44162686,44162956,44163074,44163076,44163119,44163189,44163667,44163825,44164068,44165089,44165281,44165467,44165645,44165829],"created_at":"2025-06-02T20:01:49Z","created_at_i":1748894509,"num_comments":60,"objectID":"44162443","points":157,"story_id":44162443,"title":"Snowflake to buy Crunchy Data for $250M","updated_at":"2025-12-31T19:49:06Z","url":"https://www.wsj.com/articles/snowflake-to-buy-crunchy-data-for-250-million-233543ab"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"throwawaysflake"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Throwaway to share a cautionary tale about contract chicanery with Snowflake to protect our ID. In mid-2019, I started testing DWH providers for our small data stack. Did the free trial for on demand Snowflake after hearing how it is suited for smaller use cases. After a few months of using it, we had bills in the low hundreds range/month. Their sales rep reached out to sell us on a contract where we pre-purchase $10k credits on a yearly contract (their lowest contract at that point) where we can roll-over any un-used credits next year. I will emphasize here that the sales conversation, rep never mentioned anything regarding new purchases requirements with the roll over, and in fact portrayed it as a paperwork process to just continue the credits. The actual contract clause is pretty vague and just says unused capacity will roll over into the next contract.
Fast forward to now, we need to do our roll-over contract for our unused credits, which due to covid, we have over 90% unused credits because we had to prioritize other projects. Rep reaches out and informs us that we need to sign a new contract for $1k min. to roll-over our 9k credits and we would lose the 15% discount on the unused credits. Losing the discount is one thing, but having to purchase more just to keep what we have purchased was not at all how this roll-over process was portrayed at signing. The rep then said in fact there is a worse policy for newer customers where unused credits will only roll-over if you renew at the same price or greater only. I should count myself lucky that they had the decency to enforce this policy rather than their other arbitrary policy that is not mentioned anywhere.
If you are considering contracting with them and they are promising all these great things about discounts and roll-over, don't do it. I guess since their IPO, they no longer care about their smaller customers. There has been some chatter on HN about how Snowflake is worth so much, well...here you go."},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Do not sign a contract with Snowflake"}},"_tags":["story","author_throwawaysflake","story_25281061","ask_hn"],"author":"throwawaysflake","children":[25281288,25281326,25281344,25281345,25281347,25281374,25281382,25281434,25281491,25281547,25281629,25281688,25281689,25281690,25281735,25281785,25281808,25282082,25282158,25282307],"created_at":"2020-12-02T20:23:26Z","created_at_i":1606940606,"num_comments":39,"objectID":"25281061","points":156,"story_id":25281061,"story_text":"Throwaway to share a cautionary tale about contract chicanery with Snowflake to protect our ID. In mid-2019, I started testing DWH providers for our small data stack. Did the free trial for on demand Snowflake after hearing how it is suited for smaller use cases. After a few months of using it, we had bills in the low hundreds range/month. Their sales rep reached out to sell us on a contract where we pre-purchase $10k credits on a yearly contract (their lowest contract at that point) where we can roll-over any un-used credits next year. I will emphasize here that the sales conversation, rep never mentioned anything regarding new purchases requirements with the roll over, and in fact portrayed it as a paperwork process to just continue the credits. The actual contract clause is pretty vague and just says unused capacity will roll over into the next contract.
Fast forward to now, we need to do our roll-over contract for our unused credits, which due to covid, we have over 90% unused credits because we had to prioritize other projects. Rep reaches out and informs us that we need to sign a new contract for $1k min. to roll-over our 9k credits and we would lose the 15% discount on the unused credits. Losing the discount is one thing, but having to purchase more just to keep what we have purchased was not at all how this roll-over process was portrayed at signing. The rep then said in fact there is a worse policy for newer customers where unused credits will only roll-over if you renew at the same price or greater only. I should count myself lucky that they had the decency to enforce this policy rather than their other arbitrary policy that is not mentioned anywhere.
If you are considering contracting with them and they are promising all these great things about discounts and roll-over, don't do it. I guess since their IPO, they no longer care about their smaller customers. There has been some chatter on HN about how Snowflake is worth so much, well...here you go.","title":"Do not sign a contract with Snowflake","updated_at":"2025-02-03T17:50:31Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"zX41ZdbW"},"story_text":{"matchLevel":"none","matchedWords":[],"value":"I created a web page to compare different analytical databases (both self-managed and services, open-source and proprietary) on a realistic dataset.\nIt contains 20+ databases, each with installation and data loading scripts. And they can be compared to each other on a set of 43 queries, by data load time or by storage size.
There are switches to select different types of databases for comparison - for example, only MySQL compatible or PostgreSQL compatible.
If you play with the switches, many interesting details will be uncovered.
Full description: https://github.com/ClickHouse/ClickBench/blob/main/README.md"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Show HN: A benchmark for analytical databases (Snowflake, Druid, Redshift)"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://benchmark.clickhouse.com/"}},"_tags":["story","author_zX41ZdbW","story_32084571","show_hn"],"author":"zX41ZdbW","children":[32086097,32086490,32086697,32086809,32086833,32087211,32087916,32088594,32088666,32088719,32088800,32088802,32109082],"created_at":"2022-07-13T16:03:09Z","created_at_i":1657728189,"num_comments":57,"objectID":"32084571","points":137,"story_id":32084571,"story_text":"I created a web page to compare different analytical databases (both self-managed and services, open-source and proprietary) on a realistic dataset.\nIt contains 20+ databases, each with installation and data loading scripts. And they can be compared to each other on a set of 43 queries, by data load time or by storage size.
There are switches to select different types of databases for comparison - for example, only MySQL compatible or PostgreSQL compatible.
If you play with the switches, many interesting details will be uncovered.
Full description: https://github.com/ClickHouse/ClickBench/blob/main/README.md","title":"Show HN: A benchmark for analytical databases (Snowflake, Druid, Redshift)","updated_at":"2025-04-08T23:46:41Z","url":"https://benchmark.clickhouse.com/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"georgewfraser"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Redshift vs. BigQuery vs. Snowflake benchmark"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://blog.fivetran.com/warehouse-benchmark-dce9f4c529c1"}},"_tags":["story","author_georgewfraser","story_15434272"],"author":"georgewfraser","children":[15434618,15434932,15434962,15435038,15435141,15435221,15435444,15435782,15435874,15436209,15436305,15436830,15436994,15438197,15438287,15438426],"created_at":"2017-10-09T14:54:40Z","created_at_i":1507560880,"num_comments":57,"objectID":"15434272","points":137,"story_id":15434272,"title":"Redshift vs. BigQuery vs. Snowflake benchmark","updated_at":"2024-09-20T01:28:32Z","url":"https://blog.fivetran.com/warehouse-benchmark-dce9f4c529c1"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"talltofu"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Snowflake acquires Streamlit for $800M to help customers build data-based apps"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://techcrunch.com/2022/03/02/snowflake-acquires-streamlit-for-800m-to-help-customers-build-data-based-apps/"}},"_tags":["story","author_talltofu","story_30533799"],"author":"talltofu","children":[30533945,30534023,30534100,30534314,30534432,30534613,30534657,30535612,30535774,30535781,30536012,30536770,30536985,30537211,30537784,30610835],"created_at":"2022-03-02T21:59:27Z","created_at_i":1646258367,"num_comments":44,"objectID":"30533799","points":129,"story_id":30533799,"title":"Snowflake acquires Streamlit for $800M to help customers build data-based apps","updated_at":"2024-09-20T10:35:47Z","url":"https://techcrunch.com/2022/03/02/snowflake-acquires-streamlit-for-800m-to-help-customers-build-data-based-apps/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"ingve"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Project Snowflake: Non-blocking safe manual memory management in .NET"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://www.microsoft.com/en-us/research/publication/project-snowflake-non-blocking-safe-manual-memory-management-net/#"}},"_tags":["story","author_ingve","story_14865365"],"author":"ingve","children":[14867014,14867138,14867374,14868572,14869601,14870183,14871495,14872406],"created_at":"2017-07-27T14:06:17Z","created_at_i":1501164377,"num_comments":22,"objectID":"14865365","points":126,"story_id":14865365,"title":"Project Snowflake: Non-blocking safe manual memory management in .NET","updated_at":"2024-09-20T01:10:06Z","url":"https://www.microsoft.com/en-us/research/publication/project-snowflake-non-blocking-safe-manual-memory-management-net/#"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"Tomte"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"How to write a novel using the snowflake method (2013)"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://www.advancedfictionwriting.com/articles/snowflake-method/"}},"_tags":["story","author_Tomte","story_39239044"],"author":"Tomte","children":[39241827,39243014,39243288,39249246,39261747,39267655,39269356,39269586,39270679,39271463],"created_at":"2024-02-03T10:21:52Z","created_at_i":1706955712,"num_comments":47,"objectID":"39239044","points":125,"story_id":39239044,"title":"How to write a novel using the snowflake method (2013)","updated_at":"2026-07-11T05:05:35Z","url":"https://www.advancedfictionwriting.com/articles/snowflake-method/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"adamkochanowicz"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Ultra High Resolution Photos of Snowflakes"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://smosa.com/cool-ultra-high-resolution-photos-of-snowflakes-because-science/"}},"_tags":["story","author_adamkochanowicz","story_25414579"],"author":"adamkochanowicz","children":[25441974,25441975,25442008,25442043,25442046,25442056,25442119,25442125,25442132,25442146,25442189,25442198,25442349,25442461,25442532,25442554,25442570,25442649,25442692,25442867,25443307,25443641,25444091,25444351,25444494,25444523,25444906,25445228,25445314,25445481,25449071,25451498,25479316],"created_at":"2020-12-14T07:58:26Z","created_at_i":1607932706,"num_comments":88,"objectID":"25414579","points":124,"story_id":25414579,"title":"Ultra High Resolution Photos of Snowflakes","updated_at":"2025-09-01T00:24:50Z","url":"https://smosa.com/cool-ultra-high-resolution-photos-of-snowflakes-because-science/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"izyda"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"US consumer spending dashboard built on data from 50M+ cards now on Snowflake"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://app.snowflake.com/marketplace/listing/GZTSZ290BUX40/"}},"_tags":["story","author_izyda","story_38102885"],"author":"izyda","children":[38103630,38103710,38104038,38104209,38104299,38104310,38104398,38104439,38104484,38104938,38104944,38105866,38106219,38106326,38106526,38106645,38106997],"created_at":"2023-11-01T18:40:26Z","created_at_i":1698864026,"num_comments":82,"objectID":"38102885","points":120,"story_id":38102885,"title":"US consumer spending dashboard built on data from 50M+ cards now on Snowflake","updated_at":"2025-08-14T22:53:53Z","url":"https://app.snowflake.com/marketplace/listing/GZTSZ290BUX40/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"bkohlmann"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"My Semester with the Snowflakes"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://gen.medium.com/my-semester-with-the-snowflakes-888285f0e662"}},"_tags":["story","author_bkohlmann","story_21889505"],"author":"bkohlmann","children":[21889692,21889695,21889698,21889702,21889726,21889740,21889749,21889781,21889876,21893141,21897516],"created_at":"2019-12-27T05:49:45Z","created_at_i":1577425785,"num_comments":53,"objectID":"21889505","points":120,"story_id":21889505,"title":"My Semester with the Snowflakes","updated_at":"2026-02-25T23:11:19Z","url":"https://gen.medium.com/my-semester-with-the-snowflakes-888285f0e662"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"jolie"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Twitter Announces \"Snowflake\" for Unique Tweet IDs"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"http://engineering.twitter.com/2010/06/announcing-snowflake.html"}},"_tags":["story","author_jolie","story_1396667"],"author":"jolie","children":[1396735,1396740,1396747,1396766,1396782,1396863,1396867,1396930,1397123,1397213,1397544,1397726],"created_at":"2010-06-02T01:02:28Z","created_at_i":1275440548,"num_comments":38,"objectID":"1396667","points":111,"story_id":1396667,"title":"Twitter Announces \"Snowflake\" for Unique Tweet IDs","updated_at":"2026-03-15T19:29:35Z","url":"http://engineering.twitter.com/2010/06/announcing-snowflake.html"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"nbaksalyar"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Spirals, Snowflakes and Trees: Recursion in Pictures"},"url":{"matchLevel":"none","matchedWords":[],"value":"http://learn.hfm.io/fractals.html"}},"_tags":["story","author_nbaksalyar","story_25583631"],"author":"nbaksalyar","children":[25584439,25584620,25584631,25585646,25585798],"created_at":"2020-12-30T17:27:11Z","created_at_i":1609349231,"num_comments":12,"objectID":"25583631","points":106,"story_id":25583631,"title":"Spirals, Snowflakes and Trees: Recursion in Pictures","updated_at":"2025-09-27T21:27:58Z","url":"http://learn.hfm.io/fractals.html"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"tomerbd"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Snowflake: Graphical SFTP client and terminal emulator with helpful utilities"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://github.com/subhra74/snowflake"}},"_tags":["story","author_tomerbd","story_22190001"],"author":"tomerbd","children":[22190490,22190576,22190629,22190940,22191332,22191449,22191813,22192339,22192890,22193826,22195053,22199105,22204881,22208015],"created_at":"2020-01-30T09:09:49Z","created_at_i":1580375389,"num_comments":44,"objectID":"22190001","points":103,"story_id":22190001,"title":"Snowflake: Graphical SFTP client and terminal emulator with helpful utilities","updated_at":"2024-09-20T05:35:41Z","url":"https://github.com/subhra74/snowflake"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"oconnore"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Data Warehouse Benchmark: Redshift, Snowflake, Azure, Presto, BigQuery"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://fivetran.com/blog/warehouse-benchmark"}},"_tags":["story","author_oconnore","story_17952205"],"author":"oconnore","children":[17952545,17952635,17952651,17952801,17952961,17956801,17957352,17957453,17961185,17967373],"created_at":"2018-09-10T14:50:29Z","created_at_i":1536591029,"num_comments":38,"objectID":"17952205","points":93,"story_id":17952205,"title":"Data Warehouse Benchmark: Redshift, Snowflake, Azure, Presto, BigQuery","updated_at":"2024-09-20T03:00:19Z","url":"https://fivetran.com/blog/warehouse-benchmark"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"pors"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Snowflake to Avalanche: A Novel Metastable Consensus Protocol Family for Crypto"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://ipfs.io/ipfs/QmUy4jh5mGNZvLkjies1RWM4YuvJh5o2FYopNPVYwrRVGV"}},"_tags":["story","author_pors","story_17090733"],"author":"pors","children":[17091283,17091290,17091349,17091371,17091382,17091465,17091500,17091573,17091916,17092211,17092408,17092772,17098901],"created_at":"2018-05-17T12:30:41Z","created_at_i":1526560241,"num_comments":49,"objectID":"17090733","points":92,"story_id":17090733,"title":"Snowflake to Avalanche: A Novel Metastable Consensus Protocol Family for Crypto","updated_at":"2024-09-20T02:29:16Z","url":"https://ipfs.io/ipfs/QmUy4jh5mGNZvLkjies1RWM4YuvJh5o2FYopNPVYwrRVGV"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"_____k"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"The First Photographs of Snowflakes Discover the Groundbreaking Microphotography (2017)"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://www.openculture.com/2017/12/the-first-photographs-of-snowflakes.html"}},"_tags":["story","author______k","story_46319382"],"author":"_____k","children":[46384718,46384888,46384929,46384956,46384964,46385162,46385951,46387005,46387966,46387974,46388249],"created_at":"2025-12-18T21:58:50Z","created_at_i":1766095130,"num_comments":18,"objectID":"46319382","points":91,"story_id":46319382,"title":"The First Photographs of Snowflakes Discover the Groundbreaking Microphotography (2017)","updated_at":"2026-03-05T23:14:00Z","url":"https://www.openculture.com/2017/12/the-first-photographs-of-snowflakes.html"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"mhb"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"The highest-resolution photos of snowflakes (2021)"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://www.smithsonianmag.com/innovation/these-are-highest-resolution-photos-ever-taken-snowflakes-180976710/"}},"_tags":["story","author_mhb","story_38812912"],"author":"mhb","children":[38812980,38813532,38813545,38816518,38816591,38816984,38817051,38817459,38817787,38817936,38818181,38823213],"created_at":"2023-12-30T04:54:12Z","created_at_i":1703912052,"num_comments":26,"objectID":"38812912","points":89,"story_id":38812912,"title":"The highest-resolution photos of snowflakes (2021)","updated_at":"2025-08-14T22:49:06Z","url":"https://www.smithsonianmag.com/innovation/these-are-highest-resolution-photos-ever-taken-snowflakes-180976710/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"atakan_gurkan"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Highest resolution photos ever taken of snowflakes"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://www.smithsonianmag.com/innovation/these-are-highest-resolution-photos-ever-taken-snowflakes-180976710/"}},"_tags":["story","author_atakan_gurkan","story_25939169"],"author":"atakan_gurkan","children":[25939211,25946725,25947804,25948212,25948594,25949271,25949383,25950363,25958110,25966692],"created_at":"2021-01-28T07:29:00Z","created_at_i":1611818940,"num_comments":26,"objectID":"25939169","points":89,"story_id":25939169,"title":"Highest resolution photos ever taken of snowflakes","updated_at":"2024-09-20T07:49:44Z","url":"https://www.smithsonianmag.com/innovation/these-are-highest-resolution-photos-ever-taken-snowflakes-180976710/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"uvdn7"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Snowflake\u2019s response to Databricks\u2019 TPC-DS post"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://www.snowflake.com/blog/industry-benchmarks-and-competing-with-integrity/"}},"_tags":["story","author_uvdn7","story_29206959"],"author":"uvdn7","children":[29206985,29207739,29207763,29207770,29207807,29207926,29207932,29207934,29207938,29207952,29207989,29208011,29208014,29208373,29208529,29210242,29211326,29211468,29211574,29216172,29232793,29233286],"created_at":"2021-11-13T02:52:29Z","created_at_i":1636771949,"num_comments":102,"objectID":"29206959","points":80,"story_id":29206959,"title":"Snowflake\u2019s response to Databricks\u2019 TPC-DS post","updated_at":"2025-10-24T16:45:05Z","url":"https://www.snowflake.com/blog/industry-benchmarks-and-competing-with-integrity/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"bpodgursky"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Snowflake Is Down Globally"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://status.snowflake.com/"}},"_tags":["story","author_bpodgursky","story_27691997"],"author":"bpodgursky","children":[27692097,27692427,27692437,27692467,27692500,27692559,27692575,27692602,27692996,27694437],"created_at":"2021-06-30T18:35:26Z","created_at_i":1625078126,"num_comments":17,"objectID":"27691997","points":78,"story_id":27691997,"title":"Snowflake Is Down Globally","updated_at":"2024-09-20T08:57:57Z","url":"https://status.snowflake.com/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"Hooke"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"How big can a snowflake get?"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://www.npr.org/2023/12/25/1217356234/just-how-big-can-a-snowflake-get-it-depends-on-what-you-mean-by-snowflake"}},"_tags":["story","author_Hooke","story_38797629"],"author":"Hooke","children":[38798389,38811535,38811707,38811808,38812885,38812938,38813153,38813929,38814148,38814926,38815789],"created_at":"2023-12-28T19:41:08Z","created_at_i":1703792468,"num_comments":22,"objectID":"38797629","points":76,"story_id":38797629,"title":"How big can a snowflake get?","updated_at":"2025-08-14T22:49:06Z","url":"https://www.npr.org/2023/12/25/1217356234/just-how-big-can-a-snowflake-get-it-depends-on-what-you-mean-by-snowflake"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"katiey"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Who Ever Said No Two Snowflakes Were Alike?"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"http://www.nytimes.com/2016/01/23/science/who-ever-said-no-two-snowflakes-were-alike.html"}},"_tags":["story","author_katiey","story_10968647"],"author":"katiey","children":[10968859,10969080,10969288,10969328,10969354,10969549],"created_at":"2016-01-25T17:52:47Z","created_at_i":1453744367,"num_comments":29,"objectID":"10968647","points":75,"story_id":10968647,"title":"Who Ever Said No Two Snowflakes Were Alike?","updated_at":"2023-09-07T00:39:24Z","url":"http://www.nytimes.com/2016/01/23/science/who-ever-said-no-two-snowflakes-were-alike.html"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"baha_man"},"story_text":{"matchLevel":"none","matchedWords":[],"value":""},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Snowflake Server"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"http://martinfowler.com/bliki/SnowflakeServer.html"}},"_tags":["story","author_baha_man","story_4226099"],"author":"baha_man","children":[4226404,4226455,4226529,4227267,4227663],"created_at":"2012-07-10T20:48:08Z","created_at_i":1341953288,"num_comments":33,"objectID":"4226099","points":73,"story_id":4226099,"story_text":"","title":"Snowflake Server","updated_at":"2023-09-06T20:45:30Z","url":"http://martinfowler.com/bliki/SnowflakeServer.html"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"htrp"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Snowflake CEO Says Worker Merit Should Outweigh Diversity Goals"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://www.bloomberg.com/news/articles/2021-06-03/snowflake-ceo-says-worker-merit-should-outweigh-diversity-goals"}},"_tags":["story","author_htrp","story_27386944"],"author":"htrp","children":[27387032,27387119,27387125,27387159,27387190,27387211,27387213,27387219,27387222,27387316,27387324,27387349,27387373,27387423,27387453,27387600,27388778,27392116],"created_at":"2021-06-03T20:35:11Z","created_at_i":1622752511,"num_comments":114,"objectID":"27386944","points":66,"story_id":27386944,"title":"Snowflake CEO Says Worker Merit Should Outweigh Diversity Goals","updated_at":"2024-09-20T08:47:01Z","url":"https://www.bloomberg.com/news/articles/2021-06-03/snowflake-ceo-says-worker-merit-should-outweigh-diversity-goals"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"DrJokepu"},"story_text":{"matchLevel":"none","matchedWords":[],"value":"I have found the article on The Old New Thing: http://blogs.msdn.com/oldnewthing/archive/2010/01/04/9943210.aspx"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"What's Wrong With This Snowflake?"},"url":{"matchLevel":"none","matchedWords":[],"value":"http://www.npr.org/templates/story/story.php?storyId=121827582"}},"_tags":["story","author_DrJokepu","story_1030711"],"author":"DrJokepu","children":[1030793,1030829,1030865,1030888,1030893,1030894,1030979,1031031,1031281,1031357,1031430,1031459],"created_at":"2010-01-04T17:37:03Z","created_at_i":1262626623,"num_comments":30,"objectID":"1030711","points":60,"story_id":1030711,"story_text":"I have found the article on The Old New Thing: http://blogs.msdn.com/oldnewthing/archive/2010/01/04/9943210.aspx","title":"What's Wrong With This Snowflake?","updated_at":"2024-09-19T16:48:58Z","url":"http://www.npr.org/templates/story/story.php?storyId=121827582"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"jonbaer"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Snowflake Computing raises $263M for its data warehouse"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://venturebeat.com/2018/01/25/snowflake-computing-lands-whopping-263-million-for-its-data-warehouse/"}},"_tags":["story","author_jonbaer","story_16255310"],"author":"jonbaer","children":[16256662,16257221,16257423,16257552,16257917,16258156],"created_at":"2018-01-29T04:19:17Z","created_at_i":1517199557,"num_comments":32,"objectID":"16255310","points":58,"story_id":16255310,"title":"Snowflake Computing raises $263M for its data warehouse","updated_at":"2024-09-20T01:59:28Z","url":"https://venturebeat.com/2018/01/25/snowflake-computing-lands-whopping-263-million-for-its-data-warehouse/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"sahil_singla"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Hey HN! We are Baselit (https://baselit.ai/), a tool that automatically optimizes Snowflake costs. Here\u2019s a demo video: https://www.youtube.com/watch?v=Ls6VRzBQ-pQ.
Snowflake is one of the most widely used data warehouses today. It abstracts out the underlying compute infrastructure into \u201cwarehouses\u201d for the user - compute units with t-shirt sizes (X-Small, Small, Medium etc.). In general, if you want to lower your data processing costs, the only thing you can do is to just process less data (i.e. query optimization). But Snowflake\u2019s warehouse abstraction allows an extra dimension along which you can optimize - by minimizing the compute you need to process that same data (i.e. warehouse optimization). Baselit automates Snowflake warehouse optimization for you.
While we were working on another idea last year (AI for SQL generation), users frequently shared with us how Snowflake costs had become a top concern, and cost optimization was now their business priority. Every few months, they would manually look for opportunities to cut down on costs (removing workloads or optimizing queries) - a time consuming process. We decided to build a solution that could automate cost optimization, and complement the manual effort of data teams.
There are two key components of Baselit:
1. Automated agents that cut down on warehouse idle time. This happens in one of two ways: cache optimization (when to suspend a warehouse vs letting it run idle) and cluster optimization (optimal spin down of clusters).
You can easily find out how much these agents can save you. Here\u2019s a SQL query that you can run on your Snowflake, and it will calculate your savings - https://baselit.ai/docs/savings-estimate
2. Autoscaler that lets you create custom scaling policies for multi-cluster warehouses based on SLAs. Snowflake\u2019s default policies (Economy and Standard) are not cost optimal in most cases, and they don\u2019t give you any control.
One use case for Autoscaler is that it helps you efficiently merge several warehouses into one multi-cluster warehouse - with a custom scaling policy that is optimal for a particular type of workload. In Autoscaler, you can set a parameter called \u201cAllowed Queuing Time\u201d that controls how fast a new cluster should spin up. For e.g. if you want to merge transformation workloads, you might want to set a higher queuing time. Baselit will slow down the cluster spin up, ensuring all clusters are running at a high utilization, and you\u2019ll see a reduction in costs.
We\u2019ve built a bunch of other features that help in optimizing Snowflake costs: a dbt optimization feature that automatically picks the right warehouse size for dbt models through constant experimentation, a \u201ccost lineage\u201d, spend views by teams/roles/users, and automatic recommendations from scanning Snowflake metadata.
Due to the nature of our product (access to Snowflake metadata required), we haven\u2019t made Baselit self-serve yet. We invite you to run our savings query (https://baselit.ai/docs/savings-estimate) and find out your potential savings. And if you\u2019d like to know more about any of our features and get a live demo, you can book one at this link - https://calendly.com/baselit-sahil/baselit-demo
We\u2019d love to read your feedback and ideas on Snowflake optimization!"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Launch HN: Baselit (YC W23) \u2013 Automatically Reduce Snowflake Costs"}},"_tags":["story","author_sahil_singla","story_40299556","launch_hn"],"author":"sahil_singla","children":[40299829,40299846,40299886,40300171,40302000,40302057,40302110,40302527,40305092],"created_at":"2024-05-08T15:54:54Z","created_at_i":1715183694,"num_comments":30,"objectID":"40299556","points":48,"story_id":40299556,"story_text":"Hey HN! We are Baselit (https://baselit.ai/), a tool that automatically optimizes Snowflake costs. Here\u2019s a demo video: https://www.youtube.com/watch?v=Ls6VRzBQ-pQ.
Snowflake is one of the most widely used data warehouses today. It abstracts out the underlying compute infrastructure into \u201cwarehouses\u201d for the user - compute units with t-shirt sizes (X-Small, Small, Medium etc.). In general, if you want to lower your data processing costs, the only thing you can do is to just process less data (i.e. query optimization). But Snowflake\u2019s warehouse abstraction allows an extra dimension along which you can optimize - by minimizing the compute you need to process that same data (i.e. warehouse optimization). Baselit automates Snowflake warehouse optimization for you.
While we were working on another idea last year (AI for SQL generation), users frequently shared with us how Snowflake costs had become a top concern, and cost optimization was now their business priority. Every few months, they would manually look for opportunities to cut down on costs (removing workloads or optimizing queries) - a time consuming process. We decided to build a solution that could automate cost optimization, and complement the manual effort of data teams.
There are two key components of Baselit:
1. Automated agents that cut down on warehouse idle time. This happens in one of two ways: cache optimization (when to suspend a warehouse vs letting it run idle) and cluster optimization (optimal spin down of clusters).
You can easily find out how much these agents can save you. Here\u2019s a SQL query that you can run on your Snowflake, and it will calculate your savings - https://baselit.ai/docs/savings-estimate
2. Autoscaler that lets you create custom scaling policies for multi-cluster warehouses based on SLAs. Snowflake\u2019s default policies (Economy and Standard) are not cost optimal in most cases, and they don\u2019t give you any control.
One use case for Autoscaler is that it helps you efficiently merge several warehouses into one multi-cluster warehouse - with a custom scaling policy that is optimal for a particular type of workload. In Autoscaler, you can set a parameter called \u201cAllowed Queuing Time\u201d that controls how fast a new cluster should spin up. For e.g. if you want to merge transformation workloads, you might want to set a higher queuing time. Baselit will slow down the cluster spin up, ensuring all clusters are running at a high utilization, and you\u2019ll see a reduction in costs.
We\u2019ve built a bunch of other features that help in optimizing Snowflake costs: a dbt optimization feature that automatically picks the right warehouse size for dbt models through constant experimentation, a \u201ccost lineage\u201d, spend views by teams/roles/users, and automatic recommendations from scanning Snowflake metadata.
Due to the nature of our product (access to Snowflake metadata required), we haven\u2019t made Baselit self-serve yet. We invite you to run our savings query (https://baselit.ai/docs/savings-estimate) and find out your potential savings. And if you\u2019d like to know more about any of our features and get a live demo, you can book one at this link - https://calendly.com/baselit-sahil/baselit-demo
We\u2019d love to read your feedback and ideas on Snowflake optimization!","title":"Launch HN: Baselit (YC W23) \u2013 Automatically Reduce Snowflake Costs","updated_at":"2024-11-30T18:59:17Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"firasd"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"AlphaDec: A human-readable alternative to ULID/Snowflake IDs"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/firasd/alphadec"}},"_tags":["story","author_firasd","story_44676963"],"author":"firasd","children":[44676977,44680923,44681439,44706313,44707001,44707142,44707256,44707413,44707678],"created_at":"2025-07-24T22:14:28Z","created_at_i":1753395268,"num_comments":18,"objectID":"44676963","points":47,"story_id":44676963,"title":"AlphaDec: A human-readable alternative to ULID/Snowflake IDs","updated_at":"2026-05-30T13:29:57Z","url":"https://github.com/firasd/alphadec"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"the_child"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"How Snowflake Rebuilt Its AWS Stack on Azure"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://www.snowflake.com/blog/how-we-built-snowflake-on-azure/"}},"_tags":["story","author_the_child","story_18233829"],"author":"the_child","children":[18234280,18234330,18234419,18234461,18234723],"created_at":"2018-10-16T22:25:42Z","created_at_i":1539728742,"num_comments":27,"objectID":"18233829","points":46,"story_id":18233829,"title":"How Snowflake Rebuilt Its AWS Stack on Azure","updated_at":"2024-09-20T03:09:24Z","url":"https://www.snowflake.com/blog/how-we-built-snowflake-on-azure/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"the-mitr"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Koch Snowflake"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"http://www.tikalon.com/blog/blog.php?article=2023/Koch_snowflake"}},"_tags":["story","author_the-mitr","story_38187445"],"author":"the-mitr","children":[38199737,38199966,38200955,38201478,38202058,38203711],"created_at":"2023-11-08T06:42:53Z","created_at_i":1699425773,"num_comments":8,"objectID":"38187445","points":46,"story_id":38187445,"title":"Koch Snowflake","updated_at":"2025-12-06T19:06:12Z","url":"http://www.tikalon.com/blog/blog.php?article=2023/Koch_snowflake"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"shawndumas"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Snowflakes Under an Electron Microscope"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"http://www.wired.com/wiredscience/2010/12/snowflakes-by-microscope/"}},"_tags":["story","author_shawndumas","story_2052198"],"author":"shawndumas","children":[2052303,2052481,2052812],"created_at":"2010-12-30T14:09:45Z","created_at_i":1293718185,"num_comments":7,"objectID":"2052198","points":44,"story_id":2052198,"title":"Snowflakes Under an Electron Microscope","updated_at":"2024-09-19T17:24:24Z","url":"http://www.wired.com/wiredscience/2010/12/snowflakes-by-microscope/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"Brajeshwar"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Militarized snowflakes: The accidental beauty of Renaissance star forts"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://bigthink.com/strange-maps/star-forts/"}},"_tags":["story","author_Brajeshwar","story_47555814"],"author":"Brajeshwar","children":[47556305,47556711,47556870,47556943],"created_at":"2026-03-28T16:02:59Z","created_at_i":1774713779,"num_comments":4,"objectID":"47555814","points":43,"story_id":47555814,"title":"Militarized snowflakes: The accidental beauty of Renaissance star forts","updated_at":"2026-04-22T08:36:36Z","url":"https://bigthink.com/strange-maps/star-forts/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"ashishnm"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"Snowflake Metadata powered by FoundationDB"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["snowflake"],"value":"https://www.snowflake.net/how-foundationdb-powers-snowflake-metadata-forward/"}},"_tags":["story","author_ashishnm","story_16880379"],"author":"ashishnm","children":[16881543,16881878,16900829],"created_at":"2018-04-19T21:36:29Z","created_at_i":1524173789,"num_comments":8,"objectID":"16880379","points":42,"story_id":16880379,"title":"Snowflake Metadata powered by FoundationDB","updated_at":"2024-09-20T02:23:49Z","url":"https://www.snowflake.net/how-foundationdb-powers-snowflake-metadata-forward/"}],"hitsPerPage":50,"nbHits":1347,"nbPages":20,"page":0,"params":"query=Snowflake&tags=story&hitsPerPage=50&advancedSyntax=true&analyticsTags=backend","processingTimeMS":9,"processingTimingsMS":{"_request":{"roundTrip":15},"afterFetch":{"format":{"highlighting":1,"total":1}},"fetch":{"query":7,"total":8},"total":9},"query":"Snowflake","serverTimeMS":11}