The LHCb experiment at CERN has decided to optimise its physics reach by removing the first level hardware trigger for 2020 and beyond. In addition to requiring fully redesigned front-end electronics ...this design creates interesting challenges for the data-acquisition and the rest of the online computing system. Such a system can only be realized within realistic cost using as much off-the-shelf hardware as possible. Relevant technologies evolve very quickly and thus the system design is architecture-centred and tries to avoid to depend too much on specific technologies. In this paper we describe the design, the motivations for various choices and the current favoured options for the implementation, and the status of the R&D. We will cover the back-end readout, which contains the only custom-made component, the event-building, the event-filter infrastructure, and storage.
The architecture of the data acquisition system foreseen for the LHCb upgrade, to be installed by 2018, is devised to readout events trigger-less, synchronously with the LHC bunch crossing rate at 40 ...MHz. Within this approach the readout boards act as a bridge between the front-end electronics and the High Level Trigger (HLT) computing farm. The baseline design for the LHCb readout is an ATCA board requiring dedicated crates. A local area standard network protocol is implemented in the on-board FPGAs to read out the data. The alternative solution proposed here consists in building the readout boards as PCIe peripherals of the event-builder servers. The main architectural advantage is that protocol and link-technology of the event-builder can be left open until very late, to profit from the most cost-effective industry technology available at the time of the LHC LS2.
For Run 2 of the LHC, LHCb is replacing a significant part of its event filter farm with new compute nodes. For the evaluation of the best performing solution, we have developed a method to convert ...our high level trigger application into a stand-alone, bootable benchmark image. With additional instrumentation we turned it into a self-optimising benchmark which explores techniques such as late forking, NUMA balancing and optimal number of threads, i.e. it automatically optimises box-level performance. We have run this procedure on a wide range of Haswell-E CPUs and numerous other architectures from both Intel and AMD, including also the latest Intel micro-blade servers. We present results in terms of performance, power consumption, overheads and relative cost.
Dataflow Monitoring in LHCb Svantesson, D; Schwemmer, R; Liu, G ...
Journal of Physics: Conference Series,
01/2011, Letnik:
331, Številka:
2
Journal Article, Conference Proceeding
Recenzirano
Odprti dostop
The LHCb data-flow starts from the collection of event-fragments from more than 300 read-out boards at a rate of 1 MHz. These data are moved through a large switching network consisting of more than ...50 routers to an event-filter farm of up to 1500 servers. Accepted events are sent through a dedicated network to storage collection nodes which concatenate accepted events in to files and transfer them to mass-storage. At nominal conditions more than 30 million packets enter and leave the network every second. Precise monitoring of this data-flow down to the single packet counter is essential to trace rare but systematic sources of data-loss. We have developed a comprehensive monitoring framework allowing to verify the data-flow at every level using a variety of standard tools and protocols such as sFlow, SNMP and custom software based on the LHCb Experiment Control System frame-work. This paper starts from an analysis of the data-flow and the involved hardware and software layers. From this analysis it derives the architecture and finally presents the implementation of this monitoring system.
The LHCb Data Acquisition during LHC Run 1 Alessio, F; Brarda, L; Bonaccorsi, E ...
Journal of physics. Conference series,
01/2014, Letnik:
513, Številka:
1
Journal Article
Recenzirano
Odprti dostop
The LHCb Data Acquisition system reads data from over 300 read-out boards and distributes them to more than 1500 event-filter servers. It uses a simple push-protocol over Gigabit Ethernet. After ...filtering, the data is consolidated into files for permanent storage using a SAN-based storage system. Since the beginning of data-taking many lessons have been learned and the reliability and robustness of the system has been greatly improved. We report on these changes and improvements, their motivation and how we intend to develop the system for Run 2. We also will report on how we try to optimise the usage of CPU resources during the running of the LHC ("deferred triggering") and the implications on the data acquisition.
Today's computing elements for software based high level trigger processing (HLT) are based on nodes with multiple cores. Using process based parallelization to filter particle collisions from the ...LHCb experiment on such nodes leads to expensive consumption of memory and hence significant cost increase. In the following an approach is presented to both minimize the resource consumption of the filter applications and to reduce the startup time. Described is the duplication of threads and the handling of files open in read-write mode when duplicating filter processes and the possibility to bootstrap the event filter applications directly from preconfigured checkpoint files. This led to a reduced memory consumption of roughly 60% in the nodes of the LHCb HLT farm and an improved startup time of a factor 10.
The LHCb Experiment is a hadronic precision experiment at the LHC accelerator aimed at mainly studying b-physics by profiting from the large b-anti-b-production at LHC. The challenge of high trigger ...efficiency has driven the choice of a readout architecture allowing the main event filtering to be performed by a software trigger with access to all detector information on a processing farm based on commercial multi-core PCs. The readout architecture therefore features only a relatively relaxed hardware trigger with a fixed and short latency accepting events at 1 MHz out of a nominal proton collision rate of 30 MHz, and high bandwidth with event fragment assembly over Gigabit Ethernet. A fast central system performs the entire synchronization, event labelling and control of the readout, as well as event management including destination control, dynamic load balancing of the readout network and the farm, and handling of special events for calibrations and luminosity measurements. The event filter farm processes the events in parallel and reduces the physics event rate to about 2 kHz which are formatted and written to disk before transfer to the offline processing. A spy mechanism allows processing and reconstructing a fraction of the events for online quality checking. In addition a 5 Hz subset of the events are sent as express stream to offline for checking calibrations and software before launching the full offline processing on the main event stream.