Difference between revisions of "Comparing Canister Cycles vs Performance Counter"

From Internet Computer Wiki
Jump to: navigation, search
m
Line 1: Line 1:
This example clarifies the charges in terms of Canister Cycles (i.e. <code>ic0.canister_cycle_balance()</code>) and WASM Instructions (i.e. <code>ic0.performance_counter()</code>).
+
The example walks through a scenario of charging cycles that includes inter-canister calls and how the performance counter relates to the work done by canisters.
  
 
== Setup ==
 
== Setup ==

Revision as of 14:40, 24 May 2022

The example walks through a scenario of charging cycles that includes inter-canister calls and how the performance counter relates to the work done by canisters.

Setup

  1. The User sends an 1KiB Ingress Message to Canister A
  2. The Canister A calls Canister B with an 2KiB Update Call
  3. The Canister B responds to Canister A with an 300-byte reply
  4. The Canister A responds to User with an 500-byte reply

Example Cycles Flow

Naming conventions:

  • 1K == 1'000
  • 1KiB == 1'024 bytes
  • 1M == 1'000'000
  • 1MiB == 1'048'576 bytes
  • 1B == 1'000'000'000
  • 1T == 1'000'000'000'000
What is Happening Canister A
Cycles Balance
Canister A
Perf Counter
Canister B
Cycles Balance
Canister B
Perf Counter
1. Initial state 1T N/A1) 1T N/A1)
2. Canister A receives 1KiB Ingress Message -1,2M
-1KiB*2K
≈999'996M
N/A1) 1T N/A1)
3. Canister A starts execution 2) -5B/10*4
≈997'996M
0
(5B limit)
1T N/A1)
4. Canister A executes 100M WASM instructions 997'996M +100M
=100M
(5B limit)
1T N/A1)
5. Canister A sends 2KiB call to Canister B 3) -260K
-2KiB*1K
-5B/10*4
-2MiB*1K
≈993'897M
+3*21
+2KiB
≈100M
(5B limit)
1T N/A1)
5. Canister A finishes execution 4) +(5B-100M)*4/10
≈995'857M
N/A1) 1T N/A1)
6. Canister B receives a call from Canister A 995'857M N/A1) -590K
≈999'999M
N/A1)
7. Canister B starts execution 2) 995'857M N/A1) -5B/10*4
≈997'999M
0
(5B limit)
8. Canister B executes 100M WASM instructions 995'857M N/A1) 997'999M +100M
=100M
(5B limit)
9. Canister B sends 300-byte reply to Canister A 995'857M N/A1) 997'999M +2*21
+300
≈100M
(5B limit)
10. Canister B finishes execution 4) 995'857M N/A1) +(5B-100M)*4/10
≈999'959M
N/A1)
11. Canister A receives a reply from Canister B 5) +(2MiB-300)*1K
≈997'953M
N/A1) 999'959M N/A1)
12. Canister A starts reply callback execution 6) 997'953M 0
(5B limit)
999'959M N/A1)
13. Canister A executes 100M WASM instructions 997'953M +100M
=100M
(5B limit)
999'959M N/A1)
14. Canister A sends 500-byte reply to User 997'953M +2*21
+500
≈100M
(5B limit)
999'959M N/A1)
15. Canister A finishes execution 4) +(5B-100M)*4/10
≈999'913M
N/A1) 999'959M N/A1)
Total: ~87M Cycles
for ~200M WASM Instructions Executed
+ messaging
~41M Cycles
for ~100M WASM Instructions Executed
+ messaging


Notes:

  1. The Performance Counter is available only during execution, so it's marked as N/A when the execution has not started yet.
  2. At the moment, there is a 5B instructions limit per normal message execution, so the Canister get charged for the maximum amount of Instructions it can potentially execute.
  3. The Canister get charged:
    260'000
    for the inter-canister call
    2KiB * 1'000
    for byte transmission
    5B/10*4
    for 5B instructions limit to process the reply
    2MiB * 1K
    for the maximum reply size
  4. The Canister gets a refund for the unused Instructions.
  5. The Canister gets a refund for the unused reply bytes.
  6. The Canister has already been charged for the reply callback.

References

  1. The Internet Computer Interface Specification
  2. The Internet Computer Computation and Storage Costs
  3. Community Conversations | Charging Canister Cycles