This is merely a historical archive of years 2008-2021, before the migration to mailman3.
A maintained and still updated list archive can be found at https://lists.osmocom.org/hyperkitty/list/gerrit-log@lists.osmocom.org/.
neels gerrit-no-reply at lists.osmocom.orgneels has submitted this change. ( https://gerrit.osmocom.org/c/osmo-bsc/+/22085 )
Change subject: hodec2: to balance congestion, use overload percent
......................................................................
hodec2: to balance congestion, use overload percent
For balancing load across congested cells and across congested TCH/*
kinds, instead of comparing the number of lchans above the configured
congestion threshold, compare the percent of lchans of overload.
In short, using a percentage prevents cells with less min-free-slots to
fill up 100% while neighbor cells still may have several free lchans
available.
An obvious example of why this is desirable is illustrated by
test_balance_congestion_by_percentage.ho_vty:
Cell A has min-free-slots 2, and has all slots occupied.
Cell B has min-free-slots 4, and has 2 slots remaining free.
If we count congested lchans as in current master: cell A has a
congestion count of 2: two more lchans in use than "allowed". If we move
one lchan over to cell B, it ends up with a congestion count of 3, which
is worse than 2. So when counting lchans, we decide that cell A should
remain full.
Instead, when comparing percentage of remaining lchans, we would see
that cell A is loaded 100% above congestion (2 of 2 remaining lchans in
use), but when moving one lchan to cell B, that would only be 75% loaded
above its treshold (3 of 4 remaining lchans in use). So a percentage
comparison would cause a handover to cell B.
Related: SYS#5259
Change-Id: I55234c6c99eb02ceee52be0d7388bea14304930f
---
M doc/manuals/chapters/handover.adoc
M src/osmo-bsc/handover_decision_2.c
M tests/handover/test_balance_congestion_by_percentage.ho_vty
M tests/handover/test_balance_congestion_tchf_tchh.ho_vty
4 files changed, 57 insertions(+), 23 deletions(-)
Approvals:
laforge: Looks good to me, but someone else must approve
neels: Looks good to me, approved
pespin: Looks good to me, but someone else must approve
Jenkins Builder: Verified
diff --git a/doc/manuals/chapters/handover.adoc b/doc/manuals/chapters/handover.adoc
index 4de744b..56e9aea 100644
--- a/doc/manuals/chapters/handover.adoc
+++ b/doc/manuals/chapters/handover.adoc
@@ -511,9 +511,11 @@
cells. Every time, the one MS that will suffer the least RXLEV loss while still
reducing congestion will be instructed to move first.
-If a cell and its neighbors are all loaded past their `min-free-slots`
-settings, the algorithmic aim is equal load: a load-based handover will never
-cause the target cell to be more congested than the source cell.
+If a cell and its neighbors are all loaded past their `min-free-slots` settings,
+the algorithmic aim is to improve the percentage of load above the
+`min-free-slots` setting: a load-based handover always requires the target cell
+to have a lower load percentage after handover than the source cell had before
+handover.
The min-free-slots setting is a tradeoff between immediate voice service
availability and optimal reception levels. A sane choice could be:
diff --git a/src/osmo-bsc/handover_decision_2.c b/src/osmo-bsc/handover_decision_2.c
index c265f5f..0362c9b 100644
--- a/src/osmo-bsc/handover_decision_2.c
+++ b/src/osmo-bsc/handover_decision_2.c
@@ -25,6 +25,7 @@
#include <stdbool.h>
#include <errno.h>
#include <limits.h>
+#include <math.h>
#include <osmocom/bsc/debug.h>
#include <osmocom/bsc/gsm_data.h>
@@ -387,6 +388,26 @@
return false;
}
+#define LOAD_PRECISION 6
+
+/* Return a number representing overload, i.e. the fraction of lchans used above the congestion threshold.
+ * Think of it as a percentage of used lchans above congestion, just represented in a fixed-point fraction with N
+ * decimal digits of fractional part. If there is no congestion (free_tch >= min_free_tch), return 0.
+ */
+static int32_t load_above_congestion(int free_tch, int min_free_tch)
+{
+ int32_t v;
+ OSMO_ASSERT(free_tch >= 0);
+ /* Avoid division by zero when no congestion threshold is set, and return zero overload when there is no
+ * congestion. */
+ if (free_tch >= min_free_tch)
+ return 0;
+ v = min_free_tch - free_tch;
+ v *= pow(10, LOAD_PRECISION);
+ v /= min_free_tch;
+ return v;
+}
+
/*
* Check what requirements the given cell fulfills.
* A bit mask of fulfilled requirements is returned.
@@ -443,7 +464,7 @@
{
uint8_t requirement = 0;
unsigned int penalty_time;
- int current_overbooked;
+ int32_t current_overbooked;
c->requirements = 0;
/* Requirement A */
@@ -626,17 +647,25 @@
/* Requirement C */
- /* the nr of lchans surpassing congestion on the target cell must be <= the lchans surpassing congestion on the
- * current cell _after_ handover/assignment */
- current_overbooked = c->current.min_free_tch - c->current.free_tch;
+ /* the load percentage above congestion on the target cell *after* HO must be < the load percentage above
+ * congestion on the current cell, hence the - 1 on the target. */
+ current_overbooked = load_above_congestion(c->current.free_tch, c->current.min_free_tch);
if (requirement & REQUIREMENT_A_TCHF) {
- int target_overbooked = c->target.min_free_tchf - c->target.free_tchf;
- if (target_overbooked + 1 <= current_overbooked - 1)
+ int32_t target_overbooked = load_above_congestion(c->target.free_tchf - 1, c->target.min_free_tchf);
+ LOGPHOLCHANTOBTS(c->current.lchan, c->target.bts, LOGL_DEBUG,
+ "current overbooked = %s%%, TCH/F target overbooked after HO = %s%%\n",
+ osmo_int_to_float_str_c(OTC_SELECT, current_overbooked, LOAD_PRECISION - 2),
+ osmo_int_to_float_str_c(OTC_SELECT, target_overbooked, LOAD_PRECISION - 2));
+ if (target_overbooked < current_overbooked)
requirement |= REQUIREMENT_C_TCHF;
}
if (requirement & REQUIREMENT_A_TCHH) {
- int target_overbooked = c->target.min_free_tchh - c->target.free_tchh;
- if (target_overbooked + 1 <= current_overbooked - 1)
+ int32_t target_overbooked = load_above_congestion(c->target.free_tchh - 1, c->target.min_free_tchh);
+ LOGPHOLCHANTOBTS(c->current.lchan, c->target.bts, LOGL_DEBUG,
+ "current overbooked = %s%%, TCH/H target overbooked after HO = %s%%\n",
+ osmo_int_to_float_str_c(OTC_SELECT, current_overbooked, LOAD_PRECISION - 2),
+ osmo_int_to_float_str_c(OTC_SELECT, target_overbooked, LOAD_PRECISION - 2));
+ if (target_overbooked < current_overbooked)
requirement |= REQUIREMENT_C_TCHH;
}
diff --git a/tests/handover/test_balance_congestion_by_percentage.ho_vty b/tests/handover/test_balance_congestion_by_percentage.ho_vty
index e00636e..09d2151 100644
--- a/tests/handover/test_balance_congestion_by_percentage.ho_vty
+++ b/tests/handover/test_balance_congestion_by_percentage.ho_vty
@@ -29,6 +29,7 @@
meas-rep lchan * * * * rxlev 40 rxqual 0 ta 0 neighbors 30
expect-no-chan
-# bts 0 is full, but by counting lchans above congestion, it should remain full.
+# bts 0 is full, by target_overbooked_after_ho==75% < current_overbooked_before_ho==100%, a congestion balancing to bts
+# 1 is performed.
congestion-check
-expect-no-chan
+expect-ho from lchan 0 0 1 0 to lchan 1 0 5 0
diff --git a/tests/handover/test_balance_congestion_tchf_tchh.ho_vty b/tests/handover/test_balance_congestion_tchf_tchh.ho_vty
index 7f9039f..62f07bf 100644
--- a/tests/handover/test_balance_congestion_tchf_tchh.ho_vty
+++ b/tests/handover/test_balance_congestion_tchf_tchh.ho_vty
@@ -6,46 +6,48 @@
create-bts trx-count 1 timeslots c+s4 TCH/F TCH/F TCH/F TCH/F TCH/H TCH/H TCH/H
-# both TCH/H and TCH/F have one lchan above congestion, nothing happens
+# both TCH/H and TCH/F have one lchan = 33% above congestion, nothing happens
set-ts-use trx 0 0 states * TCH/F TCH/F - - TCH/HH TCH/HH -
meas-rep lchan * * * * rxlev 10 rxqual 0 ta 0
congestion-check
expect-no-chan
-# TCH/F = +1, TCH/H = +2 above congestion. Moving a TCH/H to TCH/F would just reverse the situation to F=+2 H=+1. Nothing happens.
+# TCH/F = +1 = 33%, TCH/H = +2 = 66% above congestion.
+# Moving a TCH/H to TCH/F would just reverse the situation to F=+2=66%. Nothing happens.
set-ts-use trx 0 0 states * TCH/F TCH/F - - TCH/HH TCH/HH TCH/H-
meas-rep lchan * * * * rxlev 10 rxqual 0 ta 0
congestion-check
expect-no-chan
-# F=+1 H=+3. Balance to F=+2 H=+2
+# F=+1=33% H=+3=100%. Balance to F=+2=66% (which is < 100%) and H=+2=66%
set-ts-use trx 0 0 states * TCH/F TCH/F - - TCH/HH TCH/HH TCH/HH
meas-rep lchan * * * * rxlev 10 rxqual 0 ta 0
congestion-check
expect-ho from lchan 0 0 5 0 to lchan 0 0 3 0
-# Now the exact same thing, just with different min-free-slots settings for
-# tch/f vs tch/h
+# Now similar load percentages, just with different min-free-slots settings for tch/f vs tch/h.
network
handover2 min-free-slots tch/f 3
handover2 min-free-slots tch/h 5
-# both TCH/H and TCH/F have one lchan above congestion, nothing happens
+# TCH/F has 1/3 = 33%, TCH/H has 1/5 = 20% overload.
+# Moving one to TCH/H would mean 40% overload on TCH/H, which is above the current TCH/F of 33%.
+# Nothing happens.
set-ts-use trx 0 0 states * TCH/F TCH/F - - TCH/HH - -
meas-rep lchan * * * * rxlev 20 rxqual 0 ta 0
congestion-check
expect-no-chan
-# TCH/F = +1, TCH/H = +2 above congestion. Moving a TCH/H to TCH/F would just
-# reverse the situation to F=+2 H=+1. Nothing happens.
+# TCH/F = +1 = 33%, TCH/H = +2 = 40% above congestion. Moving a TCH/H to TCH/F would result
+# in F=+2=66%>40%. Nothing happens.
set-ts-use trx 0 0 states * TCH/F TCH/F - - TCH/HH TCH/H- -
meas-rep lchan * * * * rxlev 20 rxqual 0 ta 0
congestion-check
expect-no-chan
-# F=+1 H=+3. Balance to F=+2 H=+2
-set-ts-use trx 0 0 states * TCH/F TCH/F - - TCH/HH TCH/HH -
+# F=+1=33% H=+4=80%. Balance to F=+2=66%<80% and H=+3=60%
+set-ts-use trx 0 0 states * TCH/F TCH/F - - TCH/HH TCH/HH TCH/H-
meas-rep lchan * * * * rxlev 20 rxqual 0 ta 0
congestion-check
expect-ho from lchan 0 0 5 0 to lchan 0 0 3 0
--
To view, visit https://gerrit.osmocom.org/c/osmo-bsc/+/22085
To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings
Gerrit-Project: osmo-bsc
Gerrit-Branch: master
Gerrit-Change-Id: I55234c6c99eb02ceee52be0d7388bea14304930f
Gerrit-Change-Number: 22085
Gerrit-PatchSet: 7
Gerrit-Owner: neels <nhofmeyr at sysmocom.de>
Gerrit-Reviewer: Jenkins Builder
Gerrit-Reviewer: laforge <laforge at osmocom.org>
Gerrit-Reviewer: neels <nhofmeyr at sysmocom.de>
Gerrit-Reviewer: pespin <pespin at sysmocom.de>
Gerrit-MessageType: merged
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osmocom.org/pipermail/gerrit-log/attachments/20210114/08f4de90/attachment.htm>