Flask api性能差，如何优化

Question

这是我创建的代码，用于通过和 http api 将 sql 响应返回为 json，使用 flask 创建 api 和 pyodbc 以连接 sql db。 api 按预期工作，但是当负载增加时，api 响应速度减慢到 20-30 秒，对于单个请求它只需要 250 毫秒，有什么方法可以优化代码以在高负载期间获得更好的性能

import pyodbc
import json
import datetime
import ast
from flask import Flask, jsonify
from flask import Flask
from flask_caching import Cache
cache = Cache(config={'CACHE_TYPE': 'simple'})
# Set connection parameters
server = 'fff.database.windows.net'
database = 'fffff'
username = 'dfdfd'
password = 'dgdgdgdg$#'
port = '1433'


# Create a connection string
conn_string = f'DRIVER={{ODBC Driver 17 for SQL Server}};SERVER={server},{port};DATABASE={database};UID={username};PWD={password}'

app = Flask(__name__)



@app.route('/get_data/<Myuser>')
def get_data(Myuser):
    # Connect to the database
    conn = pyodbc.connect(conn_string)

    # Retrieve idu value
    cursor = conn.cursor()
    cursor.execute(f"SELECT idu FROM indirect.usr WHERE cn='{Myuser}'")
    row = cursor.fetchone()
    if row is not None:
        idu = row[0]
    else:
        raise Exception("No matching idu found for the specified cn.")

    # Retrieve data from various tables
    if idu is not None:
        queries = [
            ("SELECT * FROM indirect.usr WHERE idu='%s'", {}),
            ("SELECT STRING_AGG(dddDGID, ',') as dddDGID from indirect.usr_dddDGID WHERE idu='%s' GROUP BY idu", {}),
            ("SELECT STRING_AGG(dddKIM, ',') as dddKIM  from indirect.usr_dddKIM WHERE idu='%s' GROUP BY idu", {}),
            ("SELECT STRING_AGG(description, ',') as description from indirect.usr_description WHERE idu='%s' GROUP BY idu", {}),
            ("select STRING_AGG(MyOrgBrands, ',') as MyOrgBrands from [indirect].[usr_MyOrgBrands] WHERE idu='%s' GROUP BY idu", {}),
            ("select STRING_AGG(MyOrgCustomerPrograms, ',') as MyOrgCustomerPrograms from [indirect].[usr_MyOrgCustomerPrograms] WHERE idu='%s' GROUP BY idu", {}),
            ("select STRING_AGG(MyOrgDealerPrograms, ',') as MyOrgDealerPrograms from [indirect].[usr_MyOrgDealerPrograms] WHERE idu='%s' GROUP BY idu", {}),
            ("select STRING_AGG(MyOrgDealerTypes, ',') as MyOrgDealerTypes  from [indirect].[usr_MyOrgDealerTypes] WHERE idu='%s' GROUP BY idu", {}),
            ("select STRING_AGG(MyOrgDistributorPrograms, ',') as MyOrgDistributorPrograms from [indirect].[usr_MyOrgDistributorPrograms] WHERE idu='%s' GROUP BY idu", {}),
            ("select STRING_AGG(MyOrgPilotPrograms, ',') as MyOrgPilotPrograms from [indirect].[usr_MyOrgPilotPrograms] WHERE idu='%s' GROUP BY idu", {}),
            ("select STRING_AGG(MyuserAdditionalBusinessFunctionalRoles, ',') as MyuserAdditionalBusinessFunctionalRoles from [indirect].[usr_MyuserAdditionalBusinessFunctionalRoles] WHERE idu='%s' GROUP BY idu", {}),
            ("select STRING_AGG(MyuserApplicationRoles, ',') as MyuserApplicationRoles FROM [indirect].[usr_MyuserApplicationRoles] WHERE idu ='%s' GROUP BY idu", {}),
            ("select STRING_AGG(MyuserVisibilities, ',') as MyuserVisibilities from [indirect].[usr_MyuserVisibilities] WHERE idu='%s' GROUP BY idu", {}),
            ("select STRING_AGG(l, ',') as l from [indirect].[usr_l] WHERE idu='%s' GROUP BY idu", {}),
            ("select STRING_AGG(mail, ',') as mail from [indirect].[usr_mail] WHERE idu='%s' GROUP BY idu", {}),
            ("select STRING_AGG(objectClass, ',') as objectClass from [indirect].[usr_objectClass] WHERE idu='%s' GROUP BY idu", {}),
            ("select STRING_AGG(postalCode, ',') as postalCode from [indirect].[usr_postalCode] WHERE idu='%s' GROUP BY idu", {}),
            ("select STRING_AGG(st, ',') as st from [indirect].[usr_st] WHERE idu='%s' GROUP BY idu", {}),
            ("select STRING_AGG(street, ',') as street from [indirect].[usr_street] WHERE idu='%s' GROUP BY idu", {}),
            ("select STRING_AGG(telephoneNumber, ',') as telephoneNumber from [indirect].[usr_telephoneNumber] WHERE idu='%s' GROUP BY idu", {})
        ]

        results = []
        for query, default_row in queries:
            cursor.execute(query % idu)
            rows = cursor.fetchall()
            result = [dict(zip([column[0] for column in cursor.description], row)) for row in rows] if rows else [dict(zip([column[0] for column in cursor.description], [None]*len(cursor.description)))]
            results.append(result)
            #print(result)
        for result in results:
            for row in result:
                for key, value in row.items():
                    if isinstance(value, datetime.datetime):
                        row[key] = value.isoformat()

    json_data = json.dumps(results)
    #print(json_data)
    values = [item for lst in results for item in lst]
    json_data1 = json.dumps(values)
    #print(json_data1)

    data = json.loads(json_data1)

    # Merge all dictionaries into a single dictionary
    merged_dict = {}
    for dictionary in data:
        merged_dict.update(dictionary)

    # Convert merged dictionary to JSON
    merged_json = json.dumps(merged_dict)

    conn.close()