정보로 유추해서 쿼리를 대략 작성해봤습니다. 

참고요 

with tmp as (
    select
        '2023-03-21 13:30:00.000'  as e_start1,
        '2023-03-21 17:30:00.000' as e_end1,
        '2023-03-21 14:43:36.000' as e_start2,
        '2023-03-21 14:43:42.000' as e_end2
    union all
    select
        '2023-03-21 13:30:00.000'  as e_start1,
        '2023-03-21 17:30:00.000' as e_end1,
        '2023-03-21 14:43:49.000' as e_start2,
        '2023-03-21 14:43:56.000' as e_end2
)
select
    max(s) as start1,
    max(e) as end1
from
(
    select
        (ROW_NUMBER() over (order by (select 1)) -1) / 2  as grp,
        case when ROW_NUMBER() over (order by (select 1)) % 2 = 1 then tm else null end as s,
        case when ROW_NUMBER() over (order by (select 1)) % 2 = 0 then tm else null end as e
    from
    (
    select
        distinct top 100 percent
        case no when 1 then e_start1
                when 2 then e_end1
                when 3 then e_start2
                when 4 then e_end2
        end as tm
    from tmp x
    cross join (select 1 as no union all select 2 union all select 3 union all select 4) b
    order by tm
    ) x
) y
group by grp