How exchanges turn order books into distributed logs

https://quant.engineering/exchange-order-book-distributed-logs.html

by rundef • 8 months ago

136 70 8 months ago

dmurray2• 7 months ago

This article both undersells and oversells the technical challenge exchanges solve.

First, it is of course possible to apply horizontal scaling through sharding. My order on Tesla doesn't affect your order on Apple, so it's possible to run each product on its own matching engine, its own set of gateways, etc. Most exchanges don't go this far: they might have one cluster for stocks starting A-E, etc. So they don't even exhaust the benefits available from horizontal scaling, partly because this would be expensive.

On the other hand, it's not just the sequencer that has to process all these events in strict order - which might make you think it's just a matter of returning a single increasing sequence number for every request. The matching engine which sits downstream of the sequencer also has to consume all the events and apply a much more complicated algorithm: the matching algorithm described in the article as "a pure function of the log".

Components outside of that can generally be scaled more easily: for example, a gateway cares only about activity on the orders it originally received.

The article is largely correct that separating the sequencer from the matching engine allows you to recover if the latter crashes. But this may only be a theoretical benefit. Replaying and reprocessing a day's worth of messages takes a substantial fraction of the day, because the system is already operating close to its capacity. And after it crashed, you still need to figure out which customers think they got their orders executed, and allow them to cancel outstanding orders.

londons_explore• 7 months ago

Once sequencing is done, the matching algorithm can run with some parallelism.

For example, Order A and order B might interact with eachother... but they also might not. If we assume they do not, we can have them processed totally independently and in parallel, and then only if we later determine they should have interacted with each other then we throw away the results and reprocess.

It is very similar to the way speculative execution happens in CPU's. Assume something then throw away the results if your assumption was wrong.

noitpmeder5• 7 months ago

Off the cuff, id expect this leads to less improvement than you might think. The vast majority of orders, especially orders arriving in sequence close to one another, are likely on a small set of extremely liquid symbols, and usually all for prices at or near the top of the book for those symbols.

Happy to discuss more, might be off the mark... these optimizations are always very interesting in their theoretical vs actual perf impact.

blibble4• 7 months ago

My order on Tesla doesn't affect your order on Apple

not necessarily

many exchanges allow orders into one instrument to match on another

(very, very common on derivatives exchanges)

halfmatthalfcat• 7 months ago

The amount of multi-leg strategies is insane

dmurray2• 7 months ago

Yes, I was going to note that this doesn't necessarily apply on derivatives exchanges. But

a) I don't know of any exchange where this could be true for specifically Apple and Tesla, so the example is OK

b) you can still get some level of isolation, even on commodities exchanges you can't typically affect the gold book with your crude oil order (the typical case is that your order to buy oil futures in 2027 matches against someone selling oil in 2026, plus someone selling a calendar spread)

c) for exchanges that do offer this kind of functionality, one of the ways they deal with high volumes is by temporarily disabling this feature.

cgio2• 7 months ago

The title is obviously the wrong way around, exchanges turn distributed logs into order books. The distributed part is a resilience decision but not essential to the design (technically writing to a disk would give persistence with less ability to recover, or with some potential gaps in the case of failure (remember there is a sequence published on the other end too, the market data feed)). As noted in the article, the sequencer is a single-threaded, not parallelisable process. Distribution is just a configuration of that single threaded path. Parallelisation is feasible to some extent by sharding across order books themselves (dependencies between books may complicate this).

tcbawo• 7 months ago

It would not surprise me at all if the sequencing step was done via FPGA processing many network inputs at line rate with a shared monotonic clock. This would give it some amount of parallelism.

cgio2• 7 months ago

good point, sequencing is very minimal, therefore some parallelism is feasible that way, but the pipeline is not that deep, at least ideally. Of course if people are chasing nano-seconds, it may make sense.

alexpotato7• 7 months ago

At a past job (hedge fund), my role was to co-ordinate investigations into why latency may have changed when sending orders.

A couple of quants had built a random forest regression model that could take inputs like time of day, exchange, order volume etc and spit out an interval of what latency had historically been in that range.

If the latency moved outside that range, an alert would fire and then I would co-ordinate a response with the a variety of teams e.g. trading, networking, Linux etc

If we excluded changes on our side as the culprit, we would reach out to the exchange and talk to our sales rep there would might also pull in networking etc.

Some exchanges, EUREX comes to mind, were phenomenal at helping us identify issues. e.g. they once swapped out a cable that was a few feet longer than the older cable and that's why the latency increased.

One day, it's IEX, of Flash Boys fame, that triggers an alert. Nothing changed on our side so we call them. We are going back and forth with the networking engineer and then the sales rep says, in almost hushed tones:

"Look, I've worked at other exchange so I get where you are coming from in asking these questions. Problem is, b/c of our founding ethos, we are actually not allowed to track our own internal latency so we really can't help you identify the root cause. I REALLY wish it was different."

I love this story b/c HN, as a technology focused site, often thinks all problems have technical solutions but sometimes it's actually a people or process solution.

Also, incentives and "philosophy of the founders" matter a lot too.

noitpmeder5• 7 months ago

Curious what your actual role was -- sounds very interesting! Project manager? Dev? Operations specialist? E.g. were you hired into this role, and what were the requisites?

alexpotato7• 7 months ago

I was what was called "Trade Desk".

Many firms have them and they are a hybrid of:

- DevOps (e.g. we help, or own, deployments to production)

- SRE (e.g. we own the dashboards that monitored trading and would manage outages etc)

- Trading Operations (e.g. we would work with exchanges to set up connections, cancel orders etc)

My background is:

- CompSci/Economics BA

- MBA

- ~20 years of basically doing the above roles. I started supporting an in house Order Management System at a large bank and then went from there.

For more detail, here is my LinkedIn: https://www.linkedin.com/in/alex-elliott-3210352/

I also have a thread about the types of outages you see in this line of work here: https://x.com/alexpotato/status/1215876962809339904?s=20

(I have a lot of other trading/SRE related threads here: https://x.com/alexpotato/status/1212223167944478720?s=20)

reactordev• 7 months ago

All technical problems are people problems

philipov• 7 months ago

What kind of founding ethos doesn't allow tracking internal latency? Is their founding ethos "Never Admit Responsibility?"; "Never Leave A Paper Trail?"

This company's official ethical foundation is "Don't Get Caught."

WJW• 7 months ago

From the wiki about IEX: "It was founded in 2012 in order to mitigate the effects of high-frequency trading." I can see how they don't want to track internal latency as part of that, or at least not share those numbers with outsiders. That just encourages high frequency traders again.

stuxnet79• 7 months ago

The founder Brad Katsuyama talks about his background and motivation for starting the company here - https://www.youtube.com/watch?v=N9hoqFpDjVs

It might add a bit of color to this conversation.

dboreham• 7 months ago

they once swapped out a cable that was a few feet longer than the older cable and that's why the latency increased

That was not why. Possibly the cable made a difference (had an open circuit that made the NICs back down to a lower speed; noisy leading to retransmissions) but it wasn't the length per se.

thijson3• 7 months ago

It can be the length, there's about 1ns latency per foot.

thadt• 7 months ago

Well, it depends on the granularity of the time scale right? When you're measuring milliseconds, then the cable length probably isn't a thing factoring into your latency calculation.

When we're measuring time on the scale of nanoseconds then, yes, cable length is definitely something we care about and will reliably show up in measurements. In some situations, we not only care about the cable length, but also its temperature.

lopatin• 7 months ago

Can you talk a bit more about the incentives to trade latency sensitive strategies on IEX in the first place? Is it still lucrative for its liquidity despite them artificially slowing down orders? Does a meta game evolve with HFTs all working around their system, essentially making it still a HFT playground but with extra steps? Do you think their unexpected latency increase for you guys was intentional, to free the water from sharks?

alexpotato7• 7 months ago

Every modern exchange has a single logical sequencer. No matter how many gateways feed the system, all events flow into one component whose job is to assign the next sequence number. That integer defines the global timeline.

A notable edge case here is that if EVERYTHING (e.g. market data AND orders) goes through the sequencer then you can, essentially, Denial of Service to key parts of the trading flow.

e.g. one of the first exchanges to switch to a sequencer model was famous for having big market data bursts and then huge order entry delays b/c each order got stuck in the sequencer queue. In other words, the queue would be 99.99% market data with orders sprinkled in randomly.

blibble4• 7 months ago

why would market data go through the sequenced stream on an exchange?

for an exchange: market data is a projection of the order book, an observer that sits on the stream but doesn't contribute to it

and client ports have rate limits

alexpotato7• 7 months ago

B/c, by design, you want the archived stream of events to include everything.

e.g. a lot of these systems have a "replay" node that can be used by components that just restarted. You want the replay to include ALL of the messages seen so you can rebuild the state at any given point.

(There are, of course, tradeoffs to this so I'm just commenting on the "single sequencer" design philosophy)

croemer• 7 months ago

Smells of AI writing: "Timestamps aren't enough. Exchanges need a stronger ordering primitive." etc

bwfan1232• 7 months ago

Interesting comment, I "felt" the ai too in an undescribable way. What are some obvious tells ?

andrepd• 7 months ago

The incessant bullet lists and the conclusion titled "Conclusion" give it away. And above all, the complete lack of "voice". You can tell when a human is speaking and when a sanitised amorphous blob of averageness is speaking.

Nextgrid• 7 months ago

"It isn't <this>, it's <that>" and variations thereof are the biggest tell.

rundef• 7 months ago

yes, this is AI-assisted writing, I'm not hiding from it. but this isn't just copy-pasting. I still spend up to 15 hours working on each article.

genidoi3• 7 months ago

Considering that you chose to not include your name or even a HN username in the byline of the article, there is an argument to be made that you are, in fact, hiding from it.

rhodey• 7 months ago

Always fun to read about HFT. If anyone wants to learn about the Order Book data structure you can find it in JS here:

https://github.com/rhodey/limit-order-book

https://www.npmjs.com/package/limit-order-book

8cvor6j844qw_d62• 7 months ago

Very interesting. I wished to know the author. The site doesn't seem to have readily available information on the author.

hamiecod2• 7 months ago

After reading some other articles on the site, I have a feeling that it could be written by AI.

Scubabear68• 7 months ago

I wish the article had stuck with the technical topic at hand and left out the embellishment. In particular the opening piece talking about what is happening outside the exchange.

What happens outside the exchange really doesn’t matter. The ordering will not happen until it hits the exchange.

And that is why algorithmic traders want their algos in a closet as close to the exchange both physically and also in terms of network hops as possible.

hamiecod2• 7 months ago

How long can the exchanges scale their sequencer systems (which are sequential) vertically? The trading volume is only rising with time at a higher rate than the advancement of low latency tech.

bob1029• 7 months ago

In the most ideal case a sequencer can handle ~half a billion orders per second if all it's doing is assigning a number to each item. LMAX Disruptor using value types pushes 4-500 million events/s on modern hardware.

jshaqaw2• 7 months ago

This is interesting but also just hilarious at a meta level. I was a “low frequency” ie manual fundamental based hedge fund investor for many years. In general I think hft is a net benefit to liquidity when done in compliance with the text and spirit of regulations. But no real world allocation of resources is improved by having to game transactions to this level of time granularity. This is just society pouring resources down a zero sum black hole. Open to hearing contrary views of course.

shawabawa3• 7 months ago

I've been wondering if the stock market would be more efficient if trades executed only every <small time interval> instead of continuously, i.e. every 1 second an opening trade style cross book clearance happens. Orders would have to be on the book for a full interval to execute to prevent last millisecond rushes at the end of an interval

I'm probably missing some second order effects but it feels like this would mitigate the need for race to the bottom latencies and would also provide protection against fat fingered executions in that every trading algorithm would have a full second to arbitrage it

quickthrowman• 7 months ago

You could do this but the cost would be wider bid/ask spreads for all market participants. If you make it harder for market makers to hedge their position, they will collect a larger spread to account for that. A whole lot of liquidity can disappear in a second when news hits.

I’d rather have penny-wide spreads on SPY than restrict trading speed for HFTs. Providing liquidity is beneficial to everyone, even if insane amounts of money are spent by HFTs to gain an edge.

blibble4• 7 months ago

this is exactly what many dark pools do

"continuous periodic auctions"

nick0garvey• 7 months ago

Pipelined replication: the sequencer assigns a sequence number immediately and ships the event to replicas in parallel. Matching doesn't wait for the replicas to acknowledge.

How is this avoiding data loss if the lead sequencer goes down after acking but without the replica receiving the write?

HolyLampshade• 7 months ago

I’m a tad late to the party, but it’s worth providing a little context to the technical conversation.

Of the many thing trading platforms are attempting to do, the two most relevant here are the overall latency and more importantly where serialization occurs on the system.

Latency itself is only relevant as it applies to the “uncertainty” period where capital is tied up before the result of the instruction is acknowledged. Firms can only have so much capital risk, and so these moments end up being little dead periods. So long as the latency is reasonably deterministic though it’s mostly inconsequential if a platform takes 25us or 25ms to return an order acknowledgement (this is slightly more relevant in environments where there are potentially multiple venues to trade a product on, but in terms of global financial systems these environments are exceptions and not the norm). Latency is really only important when factored alongside some metric indicating a failure of business logic (failures to execute on aggressive orders or failures to cancel in time are two typical metrics)

The most important to many participants is where serialization occurs on the trading venue (what the initial portion of this blog is about; determining who was “first”). Usually this is to the tune of 1-2ns (in some cases lower). There are diminishing returns however to making this absolute in physical terms. A small handful of venues have attempted to address serialization at the very edge of their systems, but the net result is just a change in how firms that are extremely sensitive to being first apply technical expertise to the problem.

Most “good” venues permit an amount of slop in their systems (usually to the tune of 5-10% of the overall latency) which reduces the benefits of playing the sorts of ridiculous games to be “first”. There ends up being a hard limit to the economic benefit of throwing man hours and infrastructure at the problem.

thijson3• 7 months ago

The article says it's not enough to accurately timestamp orders at the various order entry portals. I didn't understand why that's not enough.

GPS can provide fairly accurate timestamps. There's a few other GLONASS systems as well for extra reliability.

SOTGO• 7 months ago

It's probably possible to use timestamps, but I suppose you would have to handle ties in more places, with sequence numbers you only break ties once. It appears that the FIX specifications allows up to microsecond precision, but given the volume of messages it's still likely a problem. It's also easier to work with integer sequence numbers than timestamps, but that's also a small consideration.

extraduder_ire• 7 months ago

GNSS is the generic term, GLONASS is the name for the Russian system.

thijson3• 7 months ago

Thanks for the correction. I just know the Russians and Chinese have their own systems.

teleforce• 7 months ago

This distributed logs nature of the exchanges is very much suitable for Kafka.

But for the required stringent latency, Kafka for head of line (HoL) blocking under concurrent events can be an issue though [1].

[1] What If We Could Rebuild Kafka from Scratch? (220 comments)

https://news.ycombinator.com/item?id=43790420

contingencies• 7 months ago

So many wrong statements here it's difficult to know where to start. Perhaps "Why Eventual Consistency Is Impossible in Finance" which is glaring: most of the economy runs on eventual consistency (brokers, banks, credit cards, crypto consensus).

nly• 7 months ago

These distributed sequencer solutions are for resilience, and they add a lot of latency because each node needs to do something like RAFT. Exchanges generally don't care aggressively about low latency, they care about resilience and fairness. It's the hedge funds etc looking for an edge.

One thing often missed here is that most orders, even from most hedge funds and prop trading shops, still go via broker systems. Direct Market Access is getting more common but it's often a pain in the arse from a regulatory and disclosures perspective, and means you lose out on short locate (shares that you can borrow from your broker to short sell).

"Sponsored Access", where you connect directly to an exchange but your broker monitors your activity via a drop copy, is a happy middle ground.

Surprisingly though, I've heard of at least one trading venue where going direct is slower, because the venues own risk checks are slower than the the ones implemented by at least one broker, and the broker themselves are allowed to bypass the risk checks put in place at the exchange for general DMA clients. "Direct" is clearly subject to negotiation.

I've also heard of brokers who tried to implement their gateways in FPGA, and have later shuttered the project, having gone back to relatively slow software gateways for the flexibility.

A lot of trading still happens via FIX, which is a slow ASCII protocol. Most prop shops will have aggressively optimized FIX parsers and serialisers out of necessity.

People think all trading happens in these elite, bleeding-edge hardcore sub-microsecond systems, but a lot of it is just dogshit.

Things are a bit more optimised in the derivatives space because of the insane volumes (Options trading just for US equities is easily into the petabytes of storage per year).

How exchanges turn order books into distributed logs

Related Stories