sql server – SQL Azure table partitioning ,need help to understand how to apply to an existing table

While trying to optimize access to a table on an Azure SQL Db I found this link.

Now the major issues I’ve got are on this table which has around and it’ defined as

CREATE TABLE dbo.ProductInstance
(
    Id                      int IDENTITY
        CONSTRAINT PK_ProductInstance
            PRIMARY KEY,
    ProductInstanceSeriesId int       NOT NULL
        CONSTRAINT FK_ProductInstance_ProductInstanceSeries_ProductInstanceSeriesId
            REFERENCES dbo.ProductInstanceSeries,
    Title                   nvarchar(256),
    Description             nvarchar(1024),
    StartDateTime           datetime2 NOT NULL,
    EndDateTime             datetime2 NOT NULL,
    Availability            int       NOT NULL,
    InitialAvailability     int       NOT NULL,
    GuideUserId             int
        CONSTRAINT FK_ProductInstance_User_GuideUserId
            REFERENCES dbo.[User],
    SignupUserId            int
        CONSTRAINT FK_ProductInstance_User_SignupUserId
            REFERENCES dbo.[User],
    GroupNumber             int       NOT NULL,
    IsActive                bit       NOT NULL,
    IsDeleted               bit       NOT NULL,
    IsPrivate               bit       NOT NULL,
    GuideNote               nvarchar(max),
    OfficeNote              nvarchar(max),
    CreatedByUserId         int       NOT NULL,
    CreatedOnUtc            datetime2 NOT NULL,
    LastModifiedOnUtc       datetime2,
    DeletedOnUtc            datetime2,
    ProductId               int,
    MappedId                int,
    ActivityId              int,
    ActivityInstanceId      varchar(50),
    LastTimeChangedDateTime date,
    TimeChangeConfirmed     bit,
    StartedAt               datetime2,
    ProductInventoryId      int
        CONSTRAINT FK_ProductInstance_ProductInventory_ProductInventoryId
            REFERENCES dbo.ProductInventory
            ON DELETE CASCADE
)
go

Most of the searches are done on StartAt / EndAt fields so I need to understand how to partition on those fields.

In the example, it reason on a Timestamp column while in my current implementation I don’t have one…should I add one or can I reason on a Datetime2 field?

Look at the code from the link I pasted

CREATE TABLE DemoPartitionTable
(
ID int,
[Timestamp] datetime PRIMARY KEY,
[Sampletext] varchar(100)
)
ON PS_Date ([Timestamp]);
GO
INSERT INTO DemoPartitionTable (id, [TimeStamp],[SampleText])
SELECT '1', '20191231','Year 2019' UNION ALL
SELECT '2', '20211231','Year 2021' UNION ALL
SELECT '1', '20210111','Year 2021' UNION ALL
SELECT '1', '20210630','Year 2021'