Support Board
Date/Time: Wed, 28 Jan 2026 08:00:30 +0000
[User Discussion] - Denali Exchange Data Feed Latency Expectations vs Ping
View Count: 50
| [2026-01-27 21:54:27] |
| isar - Posts: 1 |
|
Hello, I’m requesting clarification on expected Denali Exchange Data Feed latency and whether my observed values are normal given my setup. Setup Summary - Data Feed: Denali Exchange Data Feed (CME) - Trading Service: Teton - Connection Mode: Direct (no VPN) - Internet Connection: Fiber 2 Gigabit/s - Sierra Chart Version: 2870 64-bit - Server Ping: ~3 ms to ds27.sierracharts.com (stable) System / SC Configuration - Running on Windows VPS: 4× Ryzen 9950 cores, 16 GB DDR5 RAM - Sierra Chart CPU usage below 3% most of the time - SC Timers set at 5 ms - Chart Update Interval: 10 ms - Intraday Data Storage Interval: 1 ms - Chartbook: single chartbook, one open chart (ESH26), no studies, all drawings off, graph draw set to blank (to measure raw latency) - Remote Buffer Delay: 20 ms, Standard Compression - NTP fully synced Logs / Heartbeat (example) 2026-01-27 15:14:20.146 | Denali Data Feed | Heartbeat from server , ServerReceivedClientHeartbeatSecondsAgo=8, NumberOfOutstandingSendBuffers=0, TransmissionDelayInMilliseconds=20, ServerSendBufferSizeInBytes=384, ActualMessageDelay=0.0 seconds, DataCompRatio=3.49, UncompressedBytes=704168, CompressionTime=0.003116(secs), NumCompressions=1603, Source=216.200.108.154, MaxSendBufferSize=8185, MaxSendBufferSizeDateTime=2026-01-27 15:13:28 Observed Behavior - Market data latency shown in Sierra Chart (Data Delay in chart header) on an ESH6 chart (empty chart, chart draw set to blank): approximately 50–90 ms, consistent across sessions (outliers to 30-40 ms as well, but most of the time it's stable at around 65-90ms) - No packet loss observed - System clock synchronized with NTP My Question: 1. Given a stable ~3 ms ping to ds27.sierracharts.com and the heartbeat showing TransmissionDelayInMilliseconds=20, is a reported Data Delay of 50-90 ms within the expected range for Denali? It's very stable around that latency, barely any latency spikes so that's very impressive and good for discretionary chart trading, but still on the high side given how fast C++ ACSIL automated trading can/could be. 2. If 40–90 ms is higher than expected, which areas should I investigate next? Possible suspects I’ve considered:- Exchange/Feed-side queuing or aggregation delay - Sierra Chart internal processing / chart update scheduling - Remote buffer delay and compression overhead - Misconfigured network settings anywhere Any other specific Sierra Chart settings, log outputs, or diagnostics I should collect and provide to help pinpoint whether this is configuration-related or inherent to the feed? I looked at all tips from the "Prices / Data Falling Behind" page (Prices / Data Falling Behind) and have implemented most of it. I can provide additional artifacts if helpful. Thank you! |
To post a message in this thread, you need to log in with your Sierra Chart account: