We can not maintain the backwards compability by giving it a minmum of 8 cores.

Especially when we provide small thin machines.

1 files changed, 2 insertions, 3 deletions

diff --git a/searchcore/src/vespa/searchcore/proton/server/proton.cpp b/searchcore/src/vespa/searchcore/proton/server/proton.cpp
index 57f36d10745..ecfca857bd5 100644
--- a/searchcore/src/vespa/searchcore/proton/server/proton.cpp
+++ b/searchcore/src/vespa/searchcore/proton/server/proton.cpp
@@ -92,11 +92,10 @@ size_t
 deriveCompactionCompressionThreads(const ProtonConfig &proton,
                                    const HwInfo::Cpu &cpuInfo) {
     size_t scaledCores = (size_t)std::ceil(cpuInfo.cores() * proton.feeding.concurrency);
-    size_t threads = std::max(scaledCores, 8ul);
 
     // We need at least 1 guaranteed free worker in order to ensure progress so #documentsdbs + 1 should suffice,
-    // but we will not be cheap and give #documentsdbs * 2
-    return std::max(threads, proton.documentdb.size() * 2);
+    // but we will not be cheap and give it one extra.
+    return std::max(scaledCores, proton.documentdb.size() + 1 + 1);
 }
 
 const vespalib::string CUSTOM_COMPONENT_API_PATH = "/state/v1/custom/component";