@@ -12,7 +12,12 @@ categories:
12
12
description : ' Out-of-order / backfilled ingestion performance considerations
13
13
14
14
'
15
- linkTitle : Out-of-order / backfilled ingestion performance considerations
15
+ linkTitle : Out-of-order/backfill performance
16
+ aliases :
17
+ - /develop/data-types/timeseries/reference/out-of-order_performance_considerations
18
+ - /develop/data-types/timeseries/reference/out-of-order_performance_considerations/
19
+ - /develop/data-types/timeseries/reference/
20
+ - /develop/data-types/timeseries/reference
16
21
title : Out-of-order / backfilled ingestion performance considerations
17
22
weight : 5
18
23
---
@@ -46,11 +51,11 @@ The graphs and tables below make these key points:
46
51
47
52
- We've observed a maximum 95% drop in the achievable ops/sec even at 99% out-of-order ingestion. (Again, reducing the chunk size can cut the impact in half.)
48
53
49
- < img src = " images/compressed-overall-ops-sec-vs-out-of-order-percentage.png " alt =" compressed-overall-ops-sec-vs-out-of-order-percentage " />
54
+ {{< image filename="/ images/timeseries/ compressed-overall-ops-sec-vs-out-of-order-percentage.webp " alt="compressed-overall-ops-sec-vs-out-of-order-percentage" >}}
50
55
51
- < img src = " images/compressed-overall-p50-lat-vs-out-of-order-percentage.png " alt =" compressed-overall-p50-lat-vs-out-of-order-percentage " />
56
+ {{< image filename="/ images/timeseries/ compressed-overall-p50-lat-vs-out-of-order-percentage.webp " alt="compressed-overall-p50-lat-vs-out-of-order-percentage" >}}
52
57
53
- < img src = " images/compressed-out-of-order-overhead-table.png " alt =" compressed-out-of-order-overhead-table " />
58
+ {{< image filename="/ images/timeseries/ compressed-out-of-order-overhead-table.webp " alt="compressed-out-of-order-overhead-table" >}}
54
59
55
60
## Uncompressed chunks out-of-order/backfilled impact analysis
56
61
@@ -63,8 +68,8 @@ Apart from that, we can observe the following key take-aways:
63
68
64
69
- We've observed a maximum 45% drop in the achievable ops/sec, even at 99% out-of-order ingestion.
65
70
66
- < img src = " images/uncompressed-overall-ops-sec-vs-out-of-order-percentage.png " alt =" uncompressed-overall-ops-sec-vs-out-of-order-percentage " />
71
+ {{< image filename="/ images/timeseries/ uncompressed-overall-ops-sec-vs-out-of-order-percentage.webp " alt="uncompressed-overall-ops-sec-vs-out-of-order-percentage" >}}
67
72
68
- < img src = " images/uncompressed-overall-p50-lat-vs-out-of-order-percentage.png " alt =" uncompressed-overall-p50-lat-vs-out-of-order-percentage " />
73
+ {{< image filename="/ images/timeseries/ uncompressed-overall-p50-lat-vs-out-of-order-percentage.webp " alt="uncompressed-overall-p50-lat-vs-out-of-order-percentage" >}}
69
74
70
- < img src = " images/uncompressed-out-of-order-overhead-table.png " alt =" uncompressed-out-of-order-overhead-table " />
75
+ {{< image filename="/ images/timeseries/ uncompressed-out-of-order-overhead-table.webp " alt="uncompressed-out-of-order-overhead-table" >}}
0 commit comments