performance - Push or Pull for a near real time automation server? -
we developing server whereby client requests interest in changes specific data elements , when data changes server pushes data client. there has vigorous debate @ work whether or not better client poll data.
what considered ideal method, in terms of performance, scalability , network load, of data transfer in near real time environment?
update: here's link gives food thought regards ui updates.
there's no ideal method every situation, push better , used more often. allows optimize server caching , data transfers, helps performance , scalability, , cuts network traffic bit avoiding client requests , empty responses. can important advantage server operate in it's own pace , supply clients data when ready.
industry standarts - such opc, gid - support both. server pushes updates subscribed clients, client can pull used data out without bothering subscription.
Comments
Post a Comment