Tailslayer: Library for reducing tail latency in RAM reads

Tailslayer: Library for reducing tail latency in RAM reads(github.com)

113 points by hasheddan 40 days ago | 23 comments

ysleepy 40 days ago |

Loved the details about how memory access actually maps addresses to channels, ranks, blocks and whatever, this is rarely discussed.

Not sure how this works for larger data structures, but my first thought was that this should be implemented as some microcode or instruction.

Most computation is not thaat jitter sensitive, perception is not really in the nano to microsecond scale, but maybe a cool gadget for like dtrace or interrupt handers etc.

shaicoleman 40 days ago |

* Announcement [1]

* Video [2]

1. https://x.com/lauriewired/status/2041566601426956391 (https://xcancel.com/lauriewired/status/2041566601426956391)

2. https://www.youtube.com/watch?v=KKbgulTp3FE

inetknght 40 days ago |

@lauriewired, I think the most interesting thing that I learned from this is that memory refresh causes readwrite stalls. For some reason I thought it was completely asynchronous.

But otherwise, nice work tying all the concepts together. You might want to get some better model trains though.

6keZbCECT2uB 40 days ago |

I like the project: taking it from refresh-induced tail latency to racing threads assigned to addresses that are de-correlated by memory channel. Connecting this to a lookup table which is broadcasted across memory channels to let the lookup paths race makes for a nice narrative, but framing this as reducing tail latency confused me because I was expecting this to do a join where a single reader gets the faster of the two racers.

From a narrative standpoint, I agree it makes more sense to focus on a duplicated lookup table and fastest wins, however, from an engineering standpoint, framing it in terms of channel de-correlated reads has more possibilities. For example, if you need to evaluate multiple parallel ML models to get a result then by intentionally partitioning your models by channel you could ensure that a model does reads on only fast data or only slow data. ML models might not be that interesting since they are good candidates for being resident in L3.

TeapotNotKettle 40 days ago |

Very interesting work.

But practically speaking, in a real application - isn’t any performance benefit going to be lost by the reduced cache hit rate caused by having a larger working set? Or are the reads of all-but-one of the replicas non-cached?

Apologies if I am missing something.

6keZbCECT2uB 40 days ago | |

Once your cache hit ratios for some data structure go < .1%, I'd rather have 75% less tail latency even if it reduces cache hit rate further.

TeapotNotKettle 40 days ago | | |

Maybe - but if that’s the case you are likely using the wrong data structure.

Additionally you are going to be memory starving every other thread/process because you are hogging all the memory channels, and making an already bad L3 cache situation worse.

Outside of extremely niche realtime use cases (which would generally fit in L3 cache) I can’t see how this would improve overall throughput, once you take into account other processes running on the same box.

Do you have an example use case?

addaon 40 days ago |

This addresses the “short long tail” (known bounded variance due to the multiple physical operations underlying a single logical memory op), but for hard real time applications the “long long tail” of correctable-ECC-error—and-scrub may be the critical case.

beached_whale 38 days ago |

I wonder if there will be a hardware solution in the future that duplicates memory over multiple channels and gives the first result back transparently without threads and racing.

Kubuxu 38 days ago | |

That somewhat existed in servers in the past, my R720xd had RAM mirroring mode. IDK if it used it for reducing latency, but you could take out a stick and the server would continue running as normal and report an alarm in iDRAC.

747fulloftapes 38 days ago | | |

No, as far as I can tell, it does not reduce latency for reads. The latency for writes is worse for both the average and worst case conditions as writes have to be sent to two dimms. The purpose is high reliability. I believe it's most analogous to RAID1 systems which, generally, only issue a read to a single device rather than taking the first succeeding of simultaneous reads.

Source: not only do I have an R720xd (and two regular R720s), I checked the Intel Xeon E5-2600v2 reference manuals.

beached_whale 38 days ago | | |

Out of my area, but yeah, I have never heard of an optimized read using that. On the surface, it seems like a task much better suited for HW and there are companies that would probably pay for the ram per core penalty to get that low jitter in latency.

jagged-chisel 40 days ago |

My understanding is that this is making a trade off of using more space to get shorter access times. Do I have that right?

OT: Tail Slayer. Not Tails Layer. My brain took longer to parse that than I’d have wanted.

thfuran 40 days ago | |

Yeah, it improves mean (but not median) access time by using more memory.

ubedan 39 days ago |

Kudos... Wonderful work that I think has real value in certain situations... Thanks for sharing!