[HN Gopher] Amazon Time Sync Service now supports microsecond-ac...
___________________________________________________________________
Amazon Time Sync Service now supports microsecond-accurate time
Author : shikhar
Score : 32 points
Date : 2023-11-17 16:34 UTC (6 hours ago)
(HTM) web link (aws.amazon.com)
(TXT) w3m dump (aws.amazon.com)
| WhatsName wrote:
| Help me out, who needs this and what are some usecases for such
| high accuracy/resolution. Genuinely curious.
| coldbrewed wrote:
| I can't comment on infrastructure in AWS, but I worked at an
| observatory for a hot minute when they used Precision Time
| Protocol (PTP). Their scheme was to record telemetry from all
| aspects of the observatory -- hardware state, environmental and
| atmospheric conditions, software and system state, commands,
| etc. The intent was to build a hyper-precise set of telemetry
| that could be used to retroactively reprocess old imaging from
| the observatory as more imaging and telemetry was generated and
| it would be easier to do that with more accurate timing.
|
| With this level of resolution, multiple observatories can more
| effectively share and synthesize data; if you know the state of
| two observatories down to the tens or hundreds of nanoseconds
| you can do more advanced work with synthetic aperture imaging.
|
| Beyond astronomy my other guess is that PTP or similar tech
| might be used for high frequency trading; it might also be
| useful for creating distributed systems; Google's Cloud Spanner
| uses TrueTime which IIRC is a proprietary equivalent of PTP.
| bestouff wrote:
| PTP is nearly always proprietary. It's very configurable and
| _has_ to be configured to work. So there are many sets of
| settings called "profiles", defined by various industries
| for their own needs. They're not completely interoperable,
| but all have in common a high precision. So you could say PTP
| is a kit to build timesync protocols.
| maxnoe wrote:
| Since this comes from a physics context, it's most likely
| talking about the white rabbit implementation of ptp, which
| is open source and open hardware:
|
| https://white-rabbit.web.cern.ch/
| jitl wrote:
| It's helpful in coordinating distributed systems, like how
| Google uses "TrueTime" in Spanner to ensure external
| consistency: https://cloud.google.com/spanner/docs/true-time-
| external-con...
|
| Garden variety NTP time / system time I wouldn't trust as an
| oracle for distributed systems but this kind of high precision
| time is probably suitable for that use.
| shikhar wrote:
| It can be used to optimize consensus and transaction protocols.
|
| Consensus example - check out Nezha
| (https://arxiv.org/abs/2206.03285), they came up with an
| interesting 'Deadline Ordered Multicast' (DOM) primitive
|
| > DOM follows Liskov's suggestion of "depending on clock
| synchronization for performance but not for correctness"
|
| Transactions example - look no further than DynamoDB,
| https://muratbuffalo.blogspot.com/2023/08/distributed-transa...
|
| > In order to relieve your anxiety before you read further, no,
| correctness does not depend on timestamps, clock
| synchronization only helps to improve performance, as more
| accurate clocks result in more successful transactions and a
| serialization order that complies with real time.
| jmgpeeter wrote:
| HFT crypto trading.
| cltby wrote:
| High frequency trading (measuring latency, inferring event
| ordering, etc). Since most crypto derivatives trading takes
| place at ap-northeast-1, it feels like AWS is orienting this
| release towards financial markets customers.
| notfish wrote:
| ...why? My napkin math says a pair of observers, one at sea level
| and one on the top of a 14,000ft mountain, will drift by about 15
| microseconds per year.
|
| In other words, the approximation that time is not relative
| breaks down at this level of accuracy - there's literally no
| value in trying to synchronize things this closely.
| perfmode wrote:
| See: Spanner
| dragontamer wrote:
| My 60-cent 32678 Hz crystal at 12.5pf is specified at +/- 20ppm
| (aka: 0.002% accuracy). But with a caveat: I need to load the
| crystal with exactly 12.5pf capacitance.
|
| So sure, I can go grab 2x 25pF capacitors and load the crystal
| to ground (series capacitors half the capacitance, so 2x 25pF
| in series == 12.5pf). Except... what?
|
| Trace-capacitance and on-chip capacitance hasn't been factored
| in.
|
| Hmmmmm... okay. Does anyone have a picoFarad accurate
| multimeter? Oh snap, those are expensive.
|
| No, the easiest way to tune a crystal oscillator to its
| appropriate accuracy is through the use of a ground truth
| clock, and comparing that ground-truth clock vs the clock. So I
| make my guess with 2x 9pF capacitors (5pF internal capacitance
| on the chip x2 pins + 2pF estimated trace capacitance) and
| measure the result against the ground truth.
|
| Well, to calibrate a 20ppm crystal oscillator means I need a
| time sync service that has 20ppm accuracy or better. That's ya
| know: +/- 20 microseconds per second accuracy.
|
| -----------
|
| That's my cheap 60-cent crystal by the way. If you go with a
| temperature-controlled oven crystal oscillator for better
| accuracy, you'll need more accuracy to tune that.
| advisedwang wrote:
| That's even more reason to sync to the microsecond level.
| Computer systems rarely need an an accurate representation of
| local time. Instead they benefit from being closely aligned
| with each other. So unavoidable drift at the microsecond level
| is a reason to sync to the microsecond level.
| sheepshear wrote:
| Correcting drift is exactly what a disciplined clock does. It's
| the solution to the problem you mentioned.
| arcastroe wrote:
| Can someone explain how this works? My laptop synchronizes time
| by performing a network request to Microsoft's servers, but my
| laptop doesn't need microsecond accuracy.
|
| In Amazon's case, a network request to synchronize time would
| take orders of magnitudes longer than microseconds. Is there
| specialized hardware involved? Or multiple requests to pin down
| the accuracy?
| DistractionRect wrote:
| Second half of the second sentence:
|
| > ... customers can now access local, GPS-disciplined reference
| clocks on supported EC2 Instances.
|
| Specialized hardware.
| grahamgooch wrote:
| High velocity trading
| fofoz wrote:
| It's a requirement to build Spanner like systems. I guess AWS is
| building something similar.
| victorbjorklund wrote:
| They could call it Dynamodb ;)
___________________________________________________________________
(page generated 2023-11-17 23:01 UTC)