Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Larry Ellison: Oracle Database 1,000x Faster Than AWS Aurora (accelerationeconomy.com)
19 points by jerryjerryjerry on June 28, 2023 | hide | past | favorite | 59 comments


They don't even bother trying to make this sound believable.

You claim it's 1.5x-2x faster? Sure I might believe that despite your prohibition against benchmarks.

10x faster? You better show how your engineers pulled out that miracle.

1000x? You've either picked out a very weird edge case, used something in a way it's not meant to be used, or just made that up. Anyone who's done any serious benchmarking in a competitive space will instantly recognize that as a 'that cant be right' result.


It's MySQL Heatwave. Pretty sure this is referring to analytics usage.

Clickhouse has a similar claim.

If you look at https://benchmark.clickhouse.com/ and pick Clickhouse and Aurora for MySQL you get something in that ballpark. Heatwave claims to be faster than both Snowflake and Clickhouse.


In that case the fairer comparison should have been to Redshift, though I will concede that Amazon walked into this one themselves by overhyping Aurora's OLAP capabilities.

OLTP != OLAP, and the sooner database vendors learn to stay in their lane and focus on one use case, the better.


Oracle claims it works for both OLTP and OLAP.

I mean on the fundamentals Heatwave adds a giant memory cache / query engine in front. That alone makes it faster than anything disk based.

As to Redshift, yes, they claim it's faster than Redshift too by quite a bit.

(I use claim as it's against the ToS to benchmark it independently)


There should not be such a distinction between in-memory and disk based databases.

Being in-memory does not make OLAP databases faster. And typically, "in-memory" is a disadvantage and a limiting factor in OLAP database architecture.

I have a presentation on this topic: https://presentations.clickhouse.com/meetup53/optimizations/

It is similar to how frequently someone says "lock-free" to highlight imaginary performance advantages. If someone says a database engine is fast because it is in-memory and lock-free, it smells wrong.


It seems like he was comparing an in-memory columnar datastore to a disk-based row-oriented database for OLAP workloads. So it certainly falls in the "used something in a way it's not meant to be used" bucket, and maybe also in the "very weird edge case" bucket if the implication is that a big company is going to keep a material chunk of its analytics data in RAM.


And of course Oracle famously forbids publishing benchmarks on their databases, so independent verification is indeed impossible.


Just wondering, is there a site hosted in a US unfriendly jurisdiction / on Tor which does publish Oracle benchmarks?


Python vs C/C++/Rust/Zig has even more startling performance differences.

If you’re looking for benchmarks of the same classes of software, among techempower benchmarks, the fastest on the charts are often well over 1000x faster.

Being 1000x slower than shit needs to be is, very literally, the default state of the vast majority of software today. I don’t even blink at 1000x claims cause that’s just the standard gains you get over popular software.


Perhaps, but Python vs C is hardly an apples-for-apples comparison. I could get 1000x differences benchmarking Aurora vs Redis for example. That's what I meant by "used in a way not meant to be used".

Broadly speaking, there's plenty of public benchmarks indicating that a properly-configured Aurora is within the same order magnitude as a properly-configured RDS Postgres which is in the same ballpark as MySQL. That's an indication to me that Aurora isn't fundamentally broken in some way, making a 1000x claim quite extraordinary.


The claim that “1000x is extraordinary” is demonstrably wrong.

You can browse virtually any modern website today and see time to full load of 15+ seconds, followed by time to link load of 7+ seconds.

Back in the days of single core pentiums, 1 second loads were considered severely broken.

Modern software on modern hardware takes tens of seconds to start, whereas software with more features in the pentium days loaded instantly.

1000x is not only not “extraordinary”, it’s actually the demonstrable norm today. You think it’s extraordinary as a result of simply not knowing what computers are capable of.


15s vs 1s is only a factor of 15x, not 1000x.

The load time of the oft-considered-bloated reddit.com for me is ~5s. Let's use that as a benchmark. If you told me you got reddit.com to load 1000x faster (5ms), then yes that is extraordinary.

You're also missing the point of my 'apples-to-apples' caveat. A modern webpage might load slowly, but it also does a lot more than a static html one-pager. Sure you might not consider that to be useful, but those features (tracking, A/B testing, ads, redundancy, CDN, etc.) are useful to someone.


And that’s why I also discussed the several web api benchmarks where you’ll see 10,000x differences from bottom ranking to top.

But you didn’t like that, so felt compelled to ignore it.

1,000x is literally nothing when discussing modern development practices.

I’m not saying that oracle is definitely 1000x faster, just that writing off claims of 1000x cause you dislike the number is not proper in software when you can immediately find 10,000x benchmarks of two directly competing software packages.


Show me a serious (i.e. used in production) web framework that is 10,000x faster or slower than React or Vue or Spring when measured on a real-life workload (i.e. not testing a single function in a hot loop).

All I'm saying is that if you're approaching a benchmark between serious competitors providing similar functionality used in the intended way, you almost never see 1000x+ performance differences.


> That's an indication to me that Aurora isn't fundamentally broken in some way, making a 1000x claim quite extraordinary.

This isn't the open source MySQL being compared. MySQL Heatwave is an in-memory analytics cluster that works as a MySQL plugin. It's cloud only.

In a way it's similar to AlloyDB from GCP, which has similar style claims (maybe not 1000x)


Maybe if you compare the fastest to the slowest. But if actually compare the fastest options in each language, you rarely see things past a 10-20x slowdown.


> Python vs C/C++/Rust/Zig has even more startling performance differences.

There's no way you're seeing a >1000x speed difference in Python/Ruby/JS and C/C++/Rust/Zig/D without building some heavily contrived comparisons. That is not a normal differential.

The "high-level" languages are slower, certainly, but usually only on a single order of magnitude.


> We use this very fast RDMA network, and we start with that. And in our Gen2 Cloud, our entire network is a super-fast network, which means that most of the applications you run in the Oracle Cloud are going to be much faster than our competitors’ clouds because they don’t use that kind of network. So we have huge cost advantages.

So instead of giving actual metrics, just say "very fast", "super-fast" and "much faster". Even for wall street that seems a bit shallow.


> We use this very fast RDMA network ...

So, Infiniband then. It's not like they're the only ones using it, nor the only ones using it at scale.

Even Azure has it as an option:

https://learn.microsoft.com/en-us/azure/virtual-machines/ove...


Funny to see this now, when thinking about a response he gave to the question 'what the hell is cloud computing', way back in ~2009:

"it is the most nonsensical .. maybe I'm an idiot, it's really just complete gibberish..."

https://www.youtube.com/watch?v=0FacYAI6DY0


That's a very misleading set of partial quotes to assemble in that order...

Ellison wasn't confused about any of the technical topics around cloud computing. He was complaining about the marketing trend at the time of slapping the "cloud computing" buzzword on all kinds of things that already existed.

This is a thing that happens in the tech world a lot, and it is pretty annoying. A more modern example is the over-use of "serverless" buzzword.


https://youtu.be/0FacYAI6DY0?t=57

Should have had an ellipsis between the 2nd and 3rd bit, but the intent wasn't altered. He's calling it gibberish and nonsense.

I wasn't trying to imply that he actually was confused, I was trying to show how he thought the term was gibberish.

I mean, who would take seriously the thought that Ellison actually thought of himself as an idiot.


No, he’s not complaining about the marketing. He’s basically saying cloud computing is nothing, just a redefinition of existing computing. He’s not complaining about the marketing. He’s saying that cloud computing is only marketing. Clearly he was wrong.


Was he clearly wrong though? I listened through the whole clip and it seems like he was spot on to me. Steve Jobs made similar comments about cloud in the 90s - it’s really just mainframe 2.0.


No, cloud at root is dynamic on-demand provisioning. A lot of times SaaS built on top of cloud was (and is) marketed as cloud, without meaningfully offering dynamic on-demand provisioning to the customer, which blurs it a bit, but cloud had a distinct definition that was neither just SaaS or “mainframe 2.0” in any sense where using the term dismissively makes any sense.


He's wrong like those "There is no cloud, it's just someone else's computer" t-shirts are wrong.

IE: not wrong, but too literal and short sighted.


> He's wrong like those "There is no cloud, it's just someone else's computer" t-shirts are wrong.

> IE: not wrong

Private cloud is cloud, and its not someone else’s computer, so, “not wrong” is not not wrong.


He's the CEO, not the CTO. I don't recall Bezos ever giving a deep-dive talk about the metrics of AWS's networking stack. He left that to his tech folks.

I have absolutely no love for Ellison as you can see in my post history, but it's a bit silly to expect detailed tech specs from the CEO of a company on an earnings call with a bunch of investors that would have no idea what he was talking about if he did.

*I did in fact forget he changed his title when he gave co-CEO to Safra Catz and Hurd to rescue Hurd and not lose Catz in what was another of a long list of ethically questionable moves. It doesn't change the fact that he's completely removed from the technical details of their products and has been for 20+ years.



he's actually the CTO.


Gives a very Trump-like feeling.


When you have an army of lawyers, they just let you sue them.


Guess which one you can't benchmark and publish results on.


It really is amazing. From Oracle's Service Agreement:

4.2 You may not, or cause or permit others to: perform or disclose any benchmark or performance tests of the Services, including the Oracle Programs, without Oracle’s prior written consent;

It's nuts!


It certainly says something about your faith in your own product when you try to contractually prevent your customers from publishing performance information about their experience.


Or that it's easy to contrive a benchmark to show anything you want, and they don't want the negative press.


Then why doesn't every other piece of software in the world ban benchmarks? Is Oracle really leading the charge into the future?


> Then why doesn't every other piece of software in the world ban benchmarks?

Lots of commercial software does. SQL Server does (or did in the past), Atlassian does, etc.


Those results could be countered or refuted by Oracle themselves or other users. This is just shady.


Gonna be a lot of negative press when your shit sucks.


Sadly this is also true of a lot of database software; there are similar clauses for DB2 and SQL Server.


My homemade database is 1,000x faster that Oracle in benchmarks.

No you can't see it. It goes to a different school.


Title is a bit misleading, since Oracle is the brand and well as the a specific database product, he is claiming that Oracles cloud hosted version of mysql is faster than AWS Aurora, not claiming that Oracle DB is faster than mysql.


That is _probably_ what he means, but there is nothing in the quotes in the article to confirm that is _actually_ the case.

Also, the comment at the end of the article "... RDMA and other key drivers behind Oracle’s ascent to being the world’s hottest major cloud vendor" demonstrates that the person who wrote the article is not interested in asking hard questions. Oracle Cloud is quite a way from being the "hottest cloud vendor" by any meaningful metric, right now.


Its mentioned in the article a few times, and in a quote from Larry.

> "We’ve announced a new database — a new version of MySQL with a fast query processor called HeatWave — and we have customers moving from Amazon Aurora where they’re experiencing a 1,000x speedup versus Aurora."


Congratulations on aggressively suing the poor souls who don't know better than to never use an Oracle product. Get lost


It would be unsurprising to find that Oracle's MySQL cloud service is 1000x faster than AWS Aurora for some specific things, or maybe even entire classes of problems.

It would be equally unsurprising if Aurora is 1000x faster for other classes of problems.


[citation needed]


larry said it?


Means nothing when you threaten academic researchers trying to benchmark your database.

Larry's word is worth as much as you think it is: nothing.


Larry, you are getting spanked by Oceanbase.

You haven't even bothered to compete since SPARC.

When you're back in the game, let us know.

  Tencent          814,854,791
  Oceanbase        707,351,007
  Oracle 11g SPARC  30,249,688
https://www.tpc.org/tpcc/results/tpcc_results5.asp?print=fal...


> we have customers moving from Amazon Aurora where they’re experiencing a 1,000x speedup versus Aurora.

> So we’re a thousand times faster in query processing than Amazon’s version of MySQL

No, “some customers are experiencing 1,000 speedup”, even if true, does not imply that, in general, your db offering is 1,000 faster.


> No, “some customers are experiencing 1,000 speedup”, even if true, does not imply that, in general, your db offering is 1,000 faster.

When you put a giant memory based parallel query engine in front, it's definitely possible to be 1000x faster :).

Heatwave adds a cluster of memory based columnar storage to MySQL. It's not your typical disk based query.


Also from a quick google search it doesn't seem HeatWave supports Serializable isolation so not all customers will be able to migrate from Aurora. It's not a 1:1 replacement in terms of functionality.


This sounds more like an attack on the AWS hardware (Nitro?) . Does Nitro not have RDMA? I'm very curious if someone can discuss this. (Currently studying for AWS Solutions architect exam)


High Performance Compute (HPC) instances on Elastic Compute (ECS) offer Remote Direct Memory Access (RDMA):

https://aws.amazon.com/about-aws/whats-new/2020/11/efa-suppo...


Was a couple of years since I was in the backend industry, but then I heard plenty of people settle for less rather than "dealing with Oracles licensing bull**".

Perhaps it's better now?


It's a trap!


What toteelal is this?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: