Does anyone have any experience of running quality CCTV video streams over a Service Provider's L3 MPLS core?
The Video (2CIF 12.5fps) would be compressed using MPEG 4 or MJPEG algorithms. I'm concerned that the video traffic would be queued together with data traffic from other Customer's who use the same DSCP markings.
In the MPLS backbone, QoS is based on EXP value which is coded on 3Bits. It means we have less class of service in the core than at the access where we rely on DSCP field usually. So yes there is great chances your video traffic shared the same class of service with other customer traffic when transported over the SP MPLS backbone.
Now it shouldn't be a concern as SP avoid congestion on their core links and they should also give you SLA's per class of service regarding the QoS service they offer to you at the access. If it matches your requirement for your video traffic, you should be fine.
You'd need to implement QOS for sure. What many people do is to have data traffic use precedence/EXP 0 and Voice precedence/EXP 5. You could make the video traffic precedence/EXP 4, for example. This also maps to certain DSCP values.
You could also consider MPLS Traffic Engineering (TE). There is point-to-point TE for your data traffic, but you could also deploy (check platform and IOS support) point-to-multipoint (p2mp) TE to carry the IP multicast (video) traffic. With TE, you could steer traffic through the network and have fast protection (FRR). You'd still need to use QOS, because MPLS TE does not automatically hook into QOS.