# US 2008/0313637 A1 · Prediction-based dynamic thread pool management method and agent platform using the same

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2008/0313637 A1
- Document ID: US20080313637A1
- Status: Published application
- Kind: A1
- Application number: 12/022,889
- Filed: 2008-01-30
- Published: 2008-12-18
- Earliest claimed: 2007-06-13
- Assignees: Sungkyunkwan University
- Inventors: Hee Yong Youn; Ji Hoon Kim
- Web page: https://patentgrep.com/patent/US20080313637A1

## CPC

- G06F 9/5027: {the resource being a machine, e.g. CPUs, Servers, Terminals}
- G06F 2209/5011: Pool
- G06F 2209/5018: Thread allocation
- G06F 2209/508: Monitor
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

The present invention relates to a prediction-based dynamic thread pool management method and an agent platform using the same. An prediction-based dynamic thread pool management method according to the present invention includes: (a) calculating a thread variation to a variation of the number of threads at a time t1, (b) calculating the number of expected threads expected at a time t2 on the basis of the thread variation, (c) determining a change of the thread variation according to the time elapsed, and (d) decreasing an executing frequency of said steps (a) and (b) when it is determined that the change of the thread variation at said step (c) is small and increasing the executing frequency of said steps (a) and (b) when it is determined that the change of the thread variation is large.

## Claims

### Claim 1

A prediction-based dynamic thread pool management method, comprising:
(a) calculating a thread variation to a variation of the number of threads at a time t1;
(b) calculating the number of expected threads expected at a time t2 on the basis of the thread variation;
(c) determining a change of the thread variation according to the time elapsed; and
(d) decreasing an executing frequency of said steps (a) and (b) when it is determined that the change of the thread variation at said step (c) is small and increasing the executing frequency of said steps (a) and (b) when it is determined that the change of the thread variation is large.

### Claim 2 (depends on 1)

The prediction-based dynamic thread pool management method according to claim 1, wherein said step (c) includes:
(c1) setting a sampling constant defining the number of the thread variations sampled per hour;
(c2) sampling the thread variations per hour as many as the sampling constant;
(c3) calculating an average and a variance of the sampled thread variations;
(c4) applying the sampled thread variations, the average of the sampled thread variations and the variance of the sampled thread variations to a Gaussian probability density function; and
(c5) determining that the change of the thread variation is small in case that the change of the thread variation is out of a predetermined reference range from a center of a result of the application of the Gaussian probability density function and determining that the change of the thread variation is large in case that the change of the thread variation is within the reference range from the center of the result of the application of the Gaussian probability density function.

### Claim 3 (depends on 2)

The prediction-based dynamic thread pool management method according to claim 2, wherein the Gaussian probability density function is represented by
f

(
x
)
=

-
(
x
-
m
)
2
/
2

σ
2
2

πσ
2
(where, x represents the thread variation, m represents the average of the thread variations, and σ2 represents the variance of the thread variations).

### Claim 4 (depends on 3)

The prediction-based dynamic thread pool management method according to claim 3, wherein said step (d) includes:
(d1) increasing the repetition number per hour of said steps (a) and (b) by a predetermined reference unit in case that it is determined that the change of the thread variation is out of the reference range from the center of the result of the application of the Gaussian probability density function in the step (c5); and
(d2) decreasing the repetition number per hour of said steps (a) and (b) by the predetermined reference unit in case that it is determined that the change of the thread variation is within the reference range from the center of the result of the application of the Gaussian probability density function in the step (c5).

### Claim 5 (depends on 4)

The prediction-based dynamic thread pool management method according to claim 4, wherein the center of the application result of the Gaussian probability density function is determined by a Q-Function.

### Claim 6 (depends on 5)

The prediction-based dynamic thread pool management method according to claim 5, wherein said steps (d1) and (d2) are represented by
{
if


∀
a
t
,
2

Q

[
a
t
-
m
σ
]
≥
R


then


Δ


t
d
=
Δ


t
d
-
Δ


t
m
if


∀
a
t
,
2

Q

[
a
t
-
m
σ
]
<
R


then


Δ


t
d
=
Δ


t
d
+
Δ


t
m
(where, at represents the thread variation, Q(x) represents the Q-Function, R represents the reference range, Δtm represents a time taken to create one thread, and Δtd represents as a value corresponding to one cycle in the executing frequency of the steps (a) and (b) and is integral multiple of Δtm).

### Claim 7 (depends on 1)

An agent platform, comprising:
a message transport protocol transmitting and receiving messages to and from agents on the basis of HTTP by using an agent communication language to enable cooperation between agents existing in heterogeneous systems; and
an agent platform core managing the agents by transmitting and receiving the messages to and from the agents via the message transport protocol by the use of a thread pool managed by a prediction-based dynamic thread pool management method according to claim 1.

### Claim 8 (depends on 7)

The agent platform according to claim 7, wherein the agent communication language includes an ACL(Agent Communication Language).

### Claim 9 (depends on 8)

The agent platform according to claim 8, wherein the agent platform core includes:
a main container being in charge of execution and management of a platform;
an agent container managing the agents by hierarchically grouping the agents with a container; and
an agent description table managing a reference value for information on the agents by using IDs of the agents as key values.

### Claim 10 (depends on 9)

The agent platform according to claim 9, wherein the agent container includes:
a life-cycle manager monitoring the conditions and operations of the agents;
an agent descriptor managing agent specification information including the IDs, addresses and functions of the agents; and
a thread pool controller managing the thread pool through the prediction-based dynamic thread pool management method.

### Claim 11 (depends on 10)

The agent platform according to claim 10, wherein the message transport protocol includes:
an HTTP communication including an HTTP server module processing messages received from the agents on the basis of HTTP and an HTTP client module processing the messages transmitted to the agents on the basis of the HTTP;
an ACL parser encoding and decoding an ACL message for transmitting and receiving data to and from the agents; and
a message queue performing ordering and queuing of the messages transmitted to and received from the agents through the HTTP communication.
