diff --git a/trieste/acquisition/optimizer.py b/trieste/acquisition/optimizer.py
index a70eb299be..c6c818428a 100644
--- a/trieste/acquisition/optimizer.py
+++ b/trieste/acquisition/optimizer.py
@@ -192,6 +192,9 @@ def generate_continuous_optimizer(
     If all `num_optimization_runs` optimizations fail to converge then we run
     `num_recovery_runs` additional runs starting from random locations (also ran in parallel).
 
+    **Note:** using a large number of `num_initial_samples` and `num_optimization_runs` with a
+    high-dimensional search space can consume a large amount of CPU memory (RAM).
+
     :param num_initial_samples: The size of the random sample used to find the starting point(s) of
         the optimization.
     :param num_optimization_runs: The number of separate optimizations to run.
diff --git a/trieste/acquisition/utils.py b/trieste/acquisition/utils.py
index 590b0bb416..8afe5d07eb 100644
--- a/trieste/acquisition/utils.py
+++ b/trieste/acquisition/utils.py
@@ -48,7 +48,8 @@ def wrapper(x: TensorType) -> TensorType:
         if length == 0:
             return fn(x)
 
-        elements_per_block = tf.size(x) / length
+        # Use int64 to calculate the input tensor size, otherwise we can overflow for large tensors.
+        elements_per_block = tf.size(x, out_type=tf.int64) / length
         blocks_per_batch = tf.cast(tf.math.ceil(split_size / elements_per_block), tf.int32)
 
         num_batches = tf.cast(tf.math.ceil(length / blocks_per_batch) - 1, tf.int32)