[ARVADOS] updated: 1.2.0-453-ga7b09c695

[Git user]

Summary of changes:
 sdk/cwl/arvados_cwl/executor.py | 26 ++++++++++++++++----------
 sdk/cwl/arvados_cwl/fsaccess.py |  2 +-
 2 files changed, 17 insertions(+), 11 deletions(-)

       via  a7b09c6956e5a98df99b7fa4322c69e502f430c6 (commit)
      from  8d8847e070f588b5d85ac2d7123fd929b4d417cd (commit)

Those revisions listed above that are new to this repository have
not appeared on any other notification email; so we list those
revisions in full, below.


commit a7b09c6956e5a98df99b7fa4322c69e502f430c6
Author: Peter Amstutz <pamstutz at veritasgenetics.com>
Date:   Wed Nov 28 14:42:06 2018 -0500

    14510: Set collection metadata cache size as 150% of sum of inputs
    
    Arvados-DCO-1.1-Signed-off-by: Peter Amstutz <pamstutz at veritasgenetics.com>

diff --git a/sdk/cwl/arvados_cwl/executor.py b/sdk/cwl/arvados_cwl/executor.py
index 29f1582aa..27e6ad41b 100644
--- a/sdk/cwl/arvados_cwl/executor.py
+++ b/sdk/cwl/arvados_cwl/executor.py
@@ -117,6 +117,7 @@ class ArvCwlExecutor(object):
         self.thread_count = arvargs.thread_count
         self.poll_interval = 12
         self.loadingContext = None
+        self.should_estimate_cache_size = True
 
         if keep_client is not None:
             self.keep_client = keep_client
@@ -125,6 +126,7 @@ class ArvCwlExecutor(object):
 
         if arvargs.collection_cache_size:
             collection_cache_size = arvargs.collection_cache_size*1024*1024
+            self.should_estimate_cache_size = False
         else:
             collection_cache_size = 256*1024*1024
 
@@ -593,16 +595,20 @@ http://doc.arvados.org/install/install-api-server.html#disable_api_methods
         if runtimeContext.priority < 1 or runtimeContext.priority > 1000:
             raise Exception("--priority must be in the range 1..1000.")
 
-        visited = set()
-        estimated_size = [0]
-        def estimate_collection_cache(obj):
-            if obj.get("location", "").startswith("keep:"):
-                m = pdh_size.match(obj["location"][5:])
-                if m and m.group(1) not in visited:
-                    visited.add(m.group(1))
-                    estimated_size[0] += int(m.group(2))
-        visit_class(job_order, ("File", "Directory"), estimate_collection_cache)
-        logger.info("Estimated size %s", estimated_size)
+        if self.should_estimate_cache_size:
+            visited = set()
+            estimated_size = [0]
+            def estimate_collection_cache(obj):
+                if obj.get("location", "").startswith("keep:"):
+                    m = pdh_size.match(obj["location"][5:])
+                    if m and m.group(1) not in visited:
+                        visited.add(m.group(1))
+                        estimated_size[0] += int(m.group(2))
+            visit_class(job_order, ("File", "Directory"), estimate_collection_cache)
+            runtimeContext.collection_cache_size = max(((estimated_size[0]*192) / (1024*1024))+1, 256)
+            self.collection_cache.set_cap(runtimeContext.collection_cache_size*1024*1024)
+
+        logger.info("Using collection cache size %s MiB", runtimeContext.collection_cache_size)
 
         runnerjob = None
         if runtimeContext.submit:
diff --git a/sdk/cwl/arvados_cwl/fsaccess.py b/sdk/cwl/arvados_cwl/fsaccess.py
index c3713b21e..0816ee8fc 100644
--- a/sdk/cwl/arvados_cwl/fsaccess.py
+++ b/sdk/cwl/arvados_cwl/fsaccess.py
@@ -53,7 +53,7 @@ class CollectionCache(object):
             if available >= required or len(self.collections) < self.min_entries:
                 return
             # cut it loose
-            logger.debug("Evicting collection reader %s from cache", pdh)
+            logger.debug("Evicting collection reader %s from cache (cap %s total %s required %s)", pdh, self.cap, self.total, required)
             del self.collections[pdh]
             self.total -= v[1]
 

-----------------------------------------------------------------------


hooks/post-receive
--