This topic provides capacity thresholds and queries per second (QPS) performance references for different specifications of cloud-native API Gateway instances to help you select the most suitable specifications for your needs.
Capacity thresholds
The following are the capacity thresholds for different gateway specifications. When gateway capacity metrics are below the alert level, you receive full Service-Level Agreement (SLA) protection. For core services, keep the gateway capacity metrics below the safe level for better stability.
-
Safe level: The gateway system maintains high throughput and low-latency performance, even if burst traffic doubles.
-
Alert level: If usage exceeds this level, gateway latency may increase and stability risks can arise during traffic bursts.
-
The apigw.dev.x1 specification is deployed on a single node and does not provide SLA protection. Use this specification for testing scenarios only. For online services, use gateway specifications that are deployed on multiple nodes.
-
SLA protection does not cover request anomalies that are caused by exceeding the gateway's CPU or memory alert levels. The gateway provides alerts for these levels. You must monitor the gateway's load level and alerts in a timely manner.
|
Gateway specifications |
Client connections |
HTTPS new connections per second |
CPU usage |
Memory usage |
||||
|
Safe level |
Alert level |
Safe level |
Alert level |
Safe level |
Alert level |
Safe level |
Alert level |
|
|
apigw.dev.x1 |
12000 |
24000 |
400 |
800 |
30% |
60% |
75% |
75% |
|
apigw.small.x1 |
24000 |
48000 |
800 |
1600 |
30% |
60% |
75% |
75% |
|
apigw.small.x2 |
48000 |
96000 |
1600 |
3200 |
30% |
60% |
75% |
75% |
|
apigw.small.x4 |
96000 |
192000 |
3200 |
6400 |
30% |
60% |
75% |
75% |
|
apigw.medium.x1 |
192000 |
384000 |
6400 |
12800 |
30% |
60% |
75% |
75% |
|
apigw.medium.x2 |
384000 |
768000 |
12800 |
25600 |
30% |
60% |
75% |
75% |
|
apigw.medium.x3 |
576000 |
1152000 |
19200 |
38400 |
30% |
60% |
75% |
75% |
|
apigw.large.x1 |
768000 |
1536000 |
25600 |
51200 |
30% |
60% |
75% |
75% |
|
apigw.large.x2 |
1536000 |
3072000 |
51200 |
102400 |
30% |
60% |
75% |
75% |
|
apigw.large.x3 |
2304000 |
4608000 |
76800 |
153600 |
30% |
60% |
75% |
75% |
|
apigw.large.x4 |
3072000 |
6144000 |
102400 |
204800 |
30% |
60% |
75% |
75% |
QPS performance reference
Gateway QPS throughput is affected by various factors, such as the response size and whether HTTPS or gzip is enabled. The following table provides reference values for the pessimistic QPS (worst-case scenario) when the gateway CPU usage is at 30%.
New HTTPS connections consume significant CPU resources. For business scenarios that involve many instantaneous HTTPS concurrent connections, refer to the data for HTTPS short-lived connections in the following table to evaluate gateway capacity.
|
Gateway specifications |
apigw.dev.x1 |
apigw.small.x1 |
apigw.small.x2 |
apigw.small.x4 |
apigw.medium.x1 |
apigw.medium.x2 |
apigw.medium.x3 |
apigw.large.x1 |
apigw.large.x2 |
apigw.large.x3 |
apigw.large.x4 |
|||
|
Connection type |
Response size (KBytes) |
Is HTTPS used? |
Use gzip compression? |
QPS reference at safe CPU level (30%) |
||||||||||
|
Short-lived connection |
1 |
No |
No |
1700 |
3400 |
6800 |
13600 |
28000 |
56000 |
84000 |
112000 |
224000 |
336000 |
448000 |
|
Yes |
No |
500 |
1000 |
2000 |
4000 |
8700 |
17400 |
26100 |
34800 |
69600 |
104400 |
139200 |
||
|
Persistent connection |
1 |
No |
No |
2200 |
4400 |
8800 |
17600 |
35000 |
70000 |
105000 |
140000 |
280000 |
420000 |
560000 |
|
Yes |
No |
2000 |
4000 |
8000 |
16000 |
32000 |
64000 |
96000 |
128000 |
256000 |
384000 |
512000 |
||
|
Yes |
Yes |
1700 |
3400 |
6800 |
13600 |
28000 |
56000 |
84000 |
112000 |
224000 |
336000 |
448000 |
||
|
10 |
No |
No |
1800 |
3600 |
7200 |
14400 |
30000 |
60000 |
90000 |
120000 |
240000 |
360000 |
480000 |
|
|
Yes |
No |
1700 |
3400 |
6800 |
13600 |
28000 |
56000 |
84000 |
112000 |
224000 |
336000 |
448000 |
||
|
Yes |
Yes |
1000 |
2000 |
4000 |
8000 |
16000 |
32000 |
48000 |
64000 |
128000 |
192000 |
256000 |
||
Quota limits
Global quotas
Global quotas are independent of gateway instance specifications. To increase a quota, submit a ticket.
|
Quota item |
Quota |
|
Instances per region |
100 |
|
API operations per API |
2000 |
Instance specification quotas
Instance specification quotas depend on the gateway instance specifications. If a quota does not meet your needs after you upgrade to a higher specification, you can submit a ticket to request an increase.
|
Dev |
Small |
Medium |
Large |
|
|
Published domain names |
100 |
200 |
500 |
1000 |
|
Associated services |
200 |
800 |
2000 |
4000 |
|
Service nodes |
400 |
1600 |
4000 |
8000 |
|
Total online routes (including HTTP API routes, REST API operations, Ingress routes) |
2500 |
5000 |
7500 |
10000 |
|
K8s service sources |
3 |
3 |
5 |
5 |
|
Installed plugins |
5 |
10 |
20 |
30 |
|
Uploaded custom plugins |
10 |
20 |
50 |
80 |
Consumer authorization count |
500 |
2000 |
6000 |
10000 |