Lines Matching refs:throughput

12   throughput high).
15 throughput. So, when needed for achieving a lower latency, BFQ builds
16 schedules that may lead to a lower throughput. If your main or only
18 throughput at all times, then do switch off all low-latency heuristics
21 latency and throughput, or on how to maximize throughput.
35 support is enabled), then the sustainable throughput with BFQ
98 High throughput
100 On hard disks, BFQ achieves up to 30% higher throughput than CFQ, and
101 up to 150% higher throughput than DEADLINE and NOOP, with all the
104 instead, about the same throughput as the other schedulers.
108 BFQ distributes the device throughput, and not just the device time,
114 applications that would otherwise cause a throughput loss.
158       throughput.
165       throughput on rotational devices and on non-queueing flash-based
168       guaranteeing the desired throughput fraction to processes
175 	guarantees the expected throughput distribution without ever
181        for throughput. So, with these devices, BFQ performs idling
184        throughput may be sub-optimal. No solution currently exists to
185        provide both strong service guarantees and optimal throughput
194     throughput. For brevity, we call just "weight-raising" the whole
203     from a high throughput. Examples are systemd during boot, or git
211     that it enables BFQ to achieve a high throughput, by queue
213     mechanism, preemption, to get a high throughput. As such EQM is a
214     unified mechanism to achieve a high throughput with interleaved
225     throughput proportional to its weight, even if the throughput
239       receive a higher fraction of the device throughput.
245       allows a high throughput to be achieved, while still providing
253 	got access to the device, the higher the throughput is.
263   guarantees the expected throughput distribution without ever idling
279 fairness) and throughput. For full details on how to choose the
280 desired tradeoff between service guarantees and throughput, see the
282 on how to maximise throughput, see slice_idle, timeout_sync and
312 throughput and making sure that the desired throughput distribution is
316 As for throughput, idling can be very helpful on highly seeky media
318 number of seeks and see improved throughput.
321 should see an overall improved throughput on faster storage devices
332 throughput and acceptable latencies.
348 where it is beneficial also for throughput, idling can severely impact
349 throughput. One important case is random workload. Because of this
351 beneficial also for throughput (as detailed in Section 2). As a
355 more important than guaranteeing maximum throughput. For example, in
357 than maximum throughput. In these cases, consider setting the
378 Setting strict_guarantees may evidently affect throughput.
428 DISABLE this mode if your only goal is to achieve a high throughput.
430 entail a lower throughput. To achieve the highest-possible throughput
440 increasing this time usually increases maximum throughput. On the
451 the algorithm, larger values increase the throughput in proportion to
461 reached a higher throughput by setting max_budget explicitly, i.e., by
499 the group lowers the throughput too much. This switch occurs on a
500 per-process basis: if a process of a leaf group causes throughput loss
530 throughput sustainable with bfq, because updating the blkio.bfq.*